cs.RO / 1 / 2607.13054
Autonomous UAV Route Planning for Coverage Maximization in Environmental Monitoring: A Systematic Literature Review
用于环境监测的自主无人机路线规划以最大化覆盖范围：系统文献综述
Abstract
Environmental monitoring with unmanned aerial vehicles (UAVs) requires route planning methods that maximize covered area while handling energy limits, operational constraints, and geometric complexity. This paper reports the protocol and preliminary results of an ongoing systematic literature review (SLR) on autonomous UAV route planning for coverage-oriented environmental monitoring. The review follows the PRISMA 2020 framework and searches Scopus and Web of Science for studies published between 2015 and 2026. The protocol focuses on path planning, coverage path planning, and informative path planning, with emphasis on algorithmic families, coverage and energy metrics, obstacle handling, geometric environment representations, and environmental constraints. At the current stage, 562 records have been identified, 161 duplicates have been removed, and 401 unique records have been screened by title, abstract, and keywords. From these, 247 studies were retained for full-text eligibility assessment (235 eligible and 12 borderline records to be resolved during full-text review). A preliminary analysis of the retained studies suggests strong concentration on coverage-oriented formulations, multi-UAV coordination, and energy-aware optimization, while fewer studies explicitly address weather, uncertainty, or obstacle-rich environments. Most retained studies rely on simulation-based validation, highlighting a potential simulation-to-reality gap, and recent publications show increasing interest in reinforcement learning, hybrid optimization, and geometry-aware planning. These early findings indicate an active but fragmented research landscape and support the need for a structured synthesis to identify mature techniques and unresolved gaps for realistic environmental monitoring missions.
Chinese Translation
使用无人机（UAV）进行环境监测需要能够在处理能量限制、操作约束和几何复杂性的同时最大化覆盖区域的路线规划方法。本文报告了一项正在进行的关于自主无人机路线规划以覆盖为导向的环境监测的系统文献综述（SLR）的协议和初步结果。该综述遵循PRISMA 2020框架，并在Scopus和Web of Science中搜索2015年至2026年间发表的研究。该协议重点关注路径规划、覆盖路径规划和信息路径规划，强调算法家族、覆盖和能量指标、障碍物处理、几何环境表示和环境约束。在当前阶段，已识别出562条记录，删除了161条重复记录，401条独特记录已通过标题、摘要和关键词进行筛选。从中保留了247项研究以进行全文合格评估（235项合格，12项边缘记录将在全文审查中解决）。对保留研究的初步分析表明，研究集中于覆盖导向的公式、多无人机协调和能量感知优化，而较少的研究明确涉及天气、不确定性或障碍物丰富的环境。大多数保留的研究依赖于基于模拟的验证，突显出潜在的模拟与现实之间的差距，最近的出版物显示出对强化学习、混合优化和几何感知规划的日益关注。这些早期发现表明，研究领域活跃但分散，支持需要进行结构化综合以识别成熟技术和未解决的差距，以便进行现实的环境监测任务。
cs.RO / 2 / 2607.13056
HRIBench: Benchmarking Interaction-Centric Human-Robot Collaboration
HRIBench：以交互为中心的人机协作基准测试
Abstract
Current vision-language-action (VLA) benchmarks primarily evaluate isolated manipulation skills while leaving human-robot interaction structure largely unmodeled. However, real-world collaboration fundamentally requires coordination under shared agency, including intent understanding, temporal synchronization, protocol adherence, and safe interaction in dynamic environments. To address this gap, we introduce HRIBench, a diagnostic benchmark for intent-aware human-robot collaboration based on executable interaction scenarios. HRIBench represents collaborative tasks as structured scenario scripts that explicitly model agent roles, temporal dependencies, coordination constraints, and human behavior distributions. Building on this abstraction, HRIBench defines three representative interaction roles: Instructor, Collaborator, and Intruder, covering intent communication, joint coordination, and robustness under human intervention. The benchmark contains 13 role-conditioned tasks with over 650 evaluation episodes generated from diverse interaction trajectories and scene variations. Beyond binary task success, HRIBench introduces interpretable interaction-centric metrics spanning synchronization, responsiveness, protocol compliance, and safety. We evaluate adapted policies based on GR00T, pi0.5, and ACT under a unified protocol. Results show that current foundation robot policies struggle substantially in collaborative settings despite strong manipulation ability, revealing major limitations in temporal coordination and intent-aware behavior. Fine-tuning on HRIBench consistently improves collaborative performance. In a real-world adaptation study, simulation data generated by HRIBench improves GR00T N1.5's physical-task success rate from 0.10 to 0.43, demonstrating the benchmark's value for advancing interaction-centric robot learning.
Chinese Translation
当前的视觉-语言-动作（VLA）基准主要评估孤立的操作技能，而在人机交互结构方面则缺乏建模。然而，现实世界的协作根本上需要在共享代理下进行协调，包括意图理解、时间同步、协议遵循以及在动态环境中的安全交互。为了解决这一问题，我们引入了HRIBench，这是一个基于可执行交互场景的意图感知人机协作的诊断基准。HRIBench将协作任务表示为结构化场景脚本，明确建模代理角色、时间依赖性、协调约束和人类行为分布。在此抽象基础上，HRIBench定义了三种代表性的交互角色：指导者（Instructor）、合作者（Collaborator）和入侵者（Intruder），涵盖意图沟通、联合协调和在人工干预下的鲁棒性。该基准包含13个角色条件任务，生成了650多个评估回合，来自多样的交互轨迹和场景变体。除了二元任务成功率外，HRIBench还引入了可解释的以交互为中心的指标，涵盖同步性、响应性、协议遵循和安全性。我们在统一协议下评估了基于GR00T、pi0.5和ACT的适应性策略。结果表明，尽管当前基础机器人策略在操作能力上表现强劲，但在协作环境中却面临重大挑战，揭示了时间协调和意图感知行为的主要局限性。在HRIBench上进行微调始终能改善协作表现。在一项现实世界的适应性研究中，由HRIBench生成的模拟数据将GR00T N1.5的物理任务成功率从0.10提高到0.43，展示了该基准在推动以交互为中心的机器人学习方面的价值。
cs.RO / 3 / 2607.13058
Power from Potential: A Survey of Electrostatic Actuators for Haptics
潜力中的动力：触觉电动执行器的综述
Abstract
As haptic interfaces integrate more seamlessly into wearables and everyday environments, they increasingly require actuators that are soft, thin, silent, and energy efficient. However, conventional motors and temperature-responsive polymers often struggle to deliver these properties due to their bulky form factors and high power consumption. High-Voltage Electrostatic Actuators (HVEAs), which generate force by applying an electric field to localized charge concentrations using high voltages and ultra-low currents, have recently emerged as a compelling alternative due to their fast, silent, and low-power operation within highly customizable and compliant form factors. This paper presents a focused review of HVEAs for haptics, examining four major classes: electrostatic switchable adhesives, dielectric elastomer actuators, soft electrohydraulic actuators, and electrokinetic pumps. For each class, we describe their mechanisms that enable haptic output; characterize their bandwidths, force densities, and spatial scalability; and evaluate their versatility for rendering cutaneous and kinesthetic feedback across wearable and world-grounded interfaces. Through this cross-technology analysis, we identify common design constraints and emerging strategies for improving ergonomics, streamlining fabrication, and integrating self-sensing. We conclude by outlining where HVEAs are uniquely positioned to advance haptic interaction and highlighting key research directions needed to translate these technologies into practical systems.
Chinese Translation
随着触觉接口在可穿戴设备和日常环境中的无缝集成，它们越来越需要柔软、薄型、安静且能效高的执行器。然而，传统电机和温度响应聚合物由于其笨重的形态和高能耗，往往难以满足这些要求。高电压静电执行器（High-Voltage Electrostatic Actuators, HVEAs）通过施加电场于局部电荷浓度，利用高电压和超低电流产生力，近年来作为一种引人注目的替代方案出现，因为它们在高度可定制和柔性形态下实现快速、安静和低功耗的操作。本文对HVEAs在触觉领域的应用进行了集中综述，考察了四大类：静电可切换粘合剂、介电弹性体执行器、软电液执行器和电动泵。对于每一类，我们描述了其实现触觉输出的机制；表征了其带宽、力密度和空间可扩展性；并评估了它们在可穿戴设备和与现实环境交互的接口中呈现皮肤感知和运动感知反馈的多样性。通过这种跨技术分析，我们识别了共同的设计约束和新兴策略，以改善人体工程学、简化制造过程和集成自感知功能。最后，我们概述了HVEAs在推动触觉交互方面的独特定位，并强调了将这些技术转化为实际系统所需的关键研究方向。
cs.RO / 4 / 2607.13059
GPUSimBench: Towards Scalable and Reliable GPU-Accelerated Simulators in Embodied AI
GPUSimBench：迈向可扩展和可靠的GPU加速模拟器在具身人工智能中的应用
Abstract
Data-driven embodied AI is rapidly transitioning into a paradigm that scales training through massively parallel simulation, where GPU-accelerated simulators serve as the foundational data infrastructure. However, as computational throughput scales, the underlying trade-offs between parallel efficiency, physical fidelity, and execution determinism remain largely unexamined, hindering the development of reliable robot learning. In this paper, we expose the hidden limits of mainstream GPU-based robotic simulators (e.g., Isaac Lab, Genesis) by introducing GPUSimBench, which focuses on scalability, physical consistency, and computational determinism. First, GPUSimBench establishes a physical grounding evaluation with a controlled inclined-plane task, quantifying the distributional alignment between simulated dynamics and their real-world counterparts. Second, we benchmark parallel scalability by measuring throughput and memory footprints across scaling environment counts. Crucially, beyond standard performance metrics, we unveil and quantify the inherent non-determinism introduced by GPU-batched execution, characterized by significant run-to-run and inter-environment variability even under identical initial conditions. Finally, we identify four empirical regimes of stochasticity within current simulator stacks, highlighting that unbounded scaling can compromise reproducibility without explicit constraints.
Chinese Translation
数据驱动的具身人工智能正在迅速转变为一种通过大规模并行模拟来扩展训练的范式，其中GPU加速的模拟器作为基础数据基础设施。然而，随着计算吞吐量的增加，平行效率、物理真实感和执行确定性之间的基本权衡仍然未得到充分研究，这阻碍了可靠机器人学习的发展。在本文中，我们通过引入GPUSimBench揭示了主流基于GPU的机器人模拟器（例如，Isaac Lab，Genesis）的隐含限制，GPUSimBench专注于可扩展性、物理一致性和计算确定性。首先，GPUSimBench建立了一个物理基础评估，采用受控的倾斜平面任务，量化模拟动态与其现实世界对应物之间的分布对齐。其次，我们通过测量在扩展环境数量下的吞吐量和内存占用来基准测试并行可扩展性。至关重要的是，除了标准性能指标外，我们揭示并量化了GPU批处理执行引入的固有非确定性，其特征是在相同初始条件下，运行到运行和环境间的变异性显著。最后，我们识别出当前模拟器堆栈中四种随机性经验范畴，强调无界扩展可能在没有明确约束的情况下损害可重复性。
cs.RO / 5 / 2607.13060
A Bayesian framework for the uncanny valley in humanoid robot design
人形机器人设计中的贝叶斯框架与不安的谷底
Abstract
The uncanny valley is a long-standing empirical rule in humanoid robot design: making robots more human-like can reduce, rather than increase, affinity. Yet existing guidelines, such as adopting robot-like appearances, avoiding excessive realism, and reducing cross-modal mismatches, remain difficult to use for algorithmic design because they are not expressed as manipulable variables. Here, we propose a hierarchical Bayesian generative model that operationalizes these guidelines as mathematical design variables. The model represents affinity toward humanoid robots as posterior-weighted negative category-conditional surprise and explains category ambiguity and perceptual mismatch as increases in surprise. It maps uncanny-valley mechanisms onto four variables: deviation from the predicted robot-category mean, inconsistency in human likeness across modalities, prediction uncertainty, and observational uncertainty. Simulations showed that category ambiguity and appearance--motion mismatch can produce affinity reductions, and that uncertainty reshapes the valley. In a human-subject experiment with robot--human morphing images, we manipulated prediction uncertainty using blurred prior robot stimuli and observational uncertainty using blurred evaluation stimuli. Increased observational uncertainty attenuated the decrease in familiarity ratings at intermediate human likeness, whereas low prediction uncertainty increased ratings for robot-like appearances. This framework turns empirical uncanny-valley heuristics into a computational basis for algorithmically evaluating and optimizing humanoid robot appearance and behavior.
Chinese Translation
不安的谷底是人形机器人设计中的一条长期存在的经验法则：使机器人更具人类特征可能会减少而不是增加亲和力。然而，现有的指导原则，如采用机器人外观、避免过度逼真和减少跨模态不匹配，因未以可操作变量的形式表达而难以用于算法设计。在此，我们提出了一种分层贝叶斯生成模型，将这些指导原则操作化为数学设计变量。该模型将对人形机器人的亲和力表示为后验加权的负类别条件惊讶，并将类别模糊性和感知不匹配解释为惊讶的增加。它将不安的谷底机制映射到四个变量上：偏离预测的机器人类别均值、跨模态的人类相似性不一致性、预测不确定性和观察不确定性。模拟结果表明，类别模糊性和外观-运动不匹配可以导致亲和力降低，而不确定性则重塑了谷底。在一项涉及机器人-人类变形图像的人类受试者实验中，我们通过模糊的先前机器人刺激操控预测不确定性，并通过模糊的评估刺激操控观察不确定性。增加的观察不确定性减轻了在中等人类相似性下熟悉度评分的下降，而低预测不确定性则提高了对机器人外观的评分。该框架将经验性的不安谷底启发式转化为一种计算基础，以算法方式评估和优化人形机器人的外观和行为。
cs.RO / 6 / 2607.13067
A 3DGS-Driven Dynamic Viewpoint and Vibrotactile Framework for Subsea Teleoperation Validated via fNIRS
基于3DGS的动态视角与振动触觉框架用于水下遥操作的验证研究
Abstract
Teleoperating remotely operated vehicles (ROVs) in flooded, cluttered infrastructure is fundamentally limited by narrow 2D egocentric views and subsea communication latency. We present a multimodal teleoperation architecture built on a ROS-Unity framework that decouples proactive spatial planning from reactive boundary avoidance. The system replaces static camera feeds with a Dynamic Adaptive Viewpoint System (DAVS), which uses continuous optimization and real-time 3D Gaussian Splatting (3DGS) to synthesize an occlusion-free exocentric viewpoint from onboard state estimation. To further reduce sensory workload, a torso-mounted vibrotactile suit maps local obstacle clearance to intuitive haptic proximity cues. The architecture was evaluated in a controlled human-subject study (N = 30) using a BlueROV2 navigating a complex simulated underwater facility. A 3 x 4 repeated-measures design compared three interaction modalities (Egocentric, Haptic, Exocentric) under four communication delays (0.0-1.0 s). Performance was quantified using behavioral measures and functional near-infrared spectroscopy (fNIRS) to assess task-evoked prefrontal activation. Results show that reactive haptic feedback improves path adherence under minimal delay, whereas the 3DGS-driven exocentric visualization provides superior resilience under severe latency (0.5-1.0 s), significantly outperforming the other modalities. fNIRS further revealed a cognitive disengagement effect: increasing latency during conventional egocentric teleoperation overloaded working memory and reduced prefrontal activation, whereas the proactive spatial context provided by DAVS sustained executive control. These findings demonstrate that spatially grounded, multimodal assistance can substantially improve operator performance and cognitive endurance during latency-degraded underwater teleoperation.
Chinese Translation
在被淹没且杂乱的基础设施中遥控操作远程操作车辆（ROVs）受到狭窄的二维自我中心视角和水下通信延迟的根本限制。我们提出了一种基于ROS-Unity框架的多模态遥操作架构，该架构将主动空间规划与反应性边界规避解耦。该系统用动态自适应视角系统（DAVS）替代静态摄像头视频流，该系统利用连续优化和实时三维高斯点云（3DGS）从车载状态估计合成无遮挡的外部视角。为了进一步减少感官负担，安装在躯干上的振动触觉套装将局部障碍物清除映射为直观的触觉接近提示。该架构在一项受控的人体实验研究中进行了评估（N = 30），使用BlueROV2在复杂的模拟水下设施中导航。采用3 x 4重复测量设计比较了三种交互模式（自我中心、触觉、外部中心）在四种通信延迟（0.0-1.0秒）下的表现。通过行为测量和功能近红外光谱（fNIRS）量化性能，以评估任务引发的前额叶激活。结果表明，在最小延迟下，反应性触觉反馈改善了路径遵循，而基于3DGS的外部中心可视化在严重延迟（0.5-1.0秒）下提供了更优的韧性，显著优于其他模式。fNIRS进一步揭示了认知脱离效应：在传统自我中心遥操作中，延迟的增加使工作记忆超负荷并降低了前额叶激活，而DAVS提供的主动空间上下文维持了执行控制。这些发现表明，基于空间的多模态辅助可以显著改善操作员在延迟降低的水下遥操作中的表现和认知耐力。
cs.RO / 7 / 2607.13072
HRO: Hierarchical Room-to-Object Framework for Zero-Shot Object Goal Navigation with Large Language Models
HRO：用于零-shot目标物体导航的大语言模型分层房间到物体框架
Abstract
Zero-shot object-goal navigation aims to enable an intelligent agent to explore and navigate to objects of unknown categories in an unfamiliar environment without specific target training. In zero-shot navigation tasks, pre-trained large models are usually employed to leverage their prior knowledge for guiding the agent's navigation. However, existing zero-shot object-goal navigation methods based on large language models (LLMs) merely utilize LLMs as flat reasoning tools to directly associate objects or regions. They lack the hierarchical spatial cognition modeling of human-like room semantics to object localization, which leads to strong blindness in exploration, insufficient accuracy in semantic association, and failure to fully unleash the common-sense reasoning potential of LLMs. This paper proposes an LLM-driven hierarchical room-to-object (HRO) framework for zero-shot object-goal navigation, which guides the agent to explore and navigate to the target object in a coarse-to-fine manner. Experiments on Gibson and HM3D datasets verify that our HRO framework achieves superior success rate and generalization over existing LLM-based methods, underscoring LLMs' strong potential for zero-shot object-goal navigation.
Chinese Translation
零-shot目标物体导航旨在使智能体能够在不进行特定目标训练的情况下，探索和导航到未知类别的物体，且环境不熟悉。在零-shot导航任务中，通常采用预训练的大模型来利用其先前知识指导智能体的导航。然而，现有基于大语言模型（LLMs）的零-shot目标物体导航方法仅将LLMs作为平面推理工具，直接关联物体或区域。它们缺乏对人类房间语义到物体定位的分层空间认知建模，这导致探索中的强盲目性、语义关联的准确性不足，以及未能充分释放LLMs的常识推理潜力。本文提出了一种基于LLM的分层房间到物体（HRO）框架，用于零-shot目标物体导航，该框架以粗到细的方式指导智能体探索和导航到目标物体。在Gibson和HM3D数据集上的实验验证了我们的HRO框架在成功率和泛化能力上优于现有基于LLM的方法，强调了LLMs在零-shot目标物体导航中的强大潜力。
cs.RO / 8 / 2607.13154
Worlds in One Demo: A Synthetic Data Engine for Learning Open-World Mobile Manipulation
一演示中的多个世界：用于学习开放世界移动操作的合成数据引擎
Abstract
Learning open-world mobile manipulation policies requires vast data to achieve spatial generalization, long-horizon robustness, and scene generalization. Current prevailing data collection paradigms, teleoperation and UMI, demand prohibitive human effort and cost at scale. To scale beyond the limits of manual data collection, we seek to maximize the value of each human demonstration by scalable data generation. To this end, we introduce WANDA: learning open-World mobile mANipulation from one demonstration via a synthetic DAta engine. WANDA first reconstructs background Gaussian splats and robot-object interaction trajectories from source RGBD observations, as a world substrate for later planning and rendering. It then rearranges contact-rich robot-object interaction segments into extensive spatial configurations, utilizing whole-body motion planning to chain them into new trajectories. To enhance long-horizon robustness, it applies Corrective State Expansion to increase the robot and object state diversity at different stages of mobile manipulation. To unlock cross-environment generalization, trajectories are synthesized on diverse generated 3D worlds from everyday photos. Furthermore, we synthesize photo-realistic observations by compositing rendered robot and object meshes with Gaussian splatting backgrounds. We evaluate our approach on extensive simulation and real-world tasks in various scenes. Experiments show that policies trained with WANDA achieve long-horizon robustness, broad spatial generalization and cross-environment generalization from one real demonstration. Moreover, WANDA naturally supports cross-embodiment data generation, validated by zero-shot deployment on another mobile manipulator with a distinct morphology.
Chinese Translation
学习开放世界移动操作策略需要大量数据，以实现空间泛化、长时间鲁棒性和场景泛化。目前主流的数据收集范式，如遥操作和用户主导的交互（UMI），在规模上需要巨大的人工努力和成本。为了超越手动数据收集的限制，我们寻求通过可扩展的数据生成来最大化每个人工演示的价值。为此，我们提出了WANDA：通过合成数据引擎从一次演示中学习开放世界移动操作。WANDA首先从源RGBD观测中重建背景高斯点云和机器人-物体交互轨迹，作为后续规划和渲染的世界基础。然后，它将富含接触的机器人-物体交互片段重新排列成广泛的空间配置，利用全身运动规划将其链接成新的轨迹。为了增强长时间鲁棒性，它应用了纠正状态扩展（Corrective State Expansion），以在移动操作的不同阶段增加机器人和物体状态的多样性。为了实现跨环境泛化，轨迹是在从日常照片生成的多样化3D世界上合成的。此外，我们通过将渲染的机器人和物体网格与高斯点云背景合成，合成了照片级真实感的观测。我们在各种场景中的广泛模拟和现实任务上评估了我们的方法。实验表明，使用WANDA训练的策略能够实现长时间鲁棒性、广泛的空间泛化和来自一次真实演示的跨环境泛化。此外，WANDA自然支持跨体现数据生成，通过在另一种形态的移动操作器上进行零样本部署得到了验证。
cs.RO / 9 / 2607.13254
Attitude Estimation Using Inertial and Barometric Measurements
基于惯性和气压测量的姿态估计
Abstract
Accurate and robust attitude estimation is a key challenge for autonomous vehicles, particularly in GNSS-denied conditions and during highly accelerated flight. In such conditions, Inertial Measurement Units (IMUs) alone are insufficient for reliable tilt estimation due to the ambiguity between gravitational and inertial accelerations. Although auxiliary velocity sensors such as GNSS, Pitot tubes, Doppler radar, or Visual Inertial Odometry are commonly used, they may be unavailable, intermittent, or costly. This paper introduces a barometer-aided attitude estimation architecture that exploits barometric altitude measurements to provide complementary information on the vehicle's vertical motion, thereby enhancing attitude estimation within nonlinear observers on SO(3). The contributions are twofold. First, we design a deterministic Riccati observer cascaded with a complementary filter, ensuring almost-global asymptotic stability (AGAS) under a uniform observability (UO) condition while preserving the geometric structure of the attitude dynamics. Second, we propose a nonlinear observer evolving on SO(3)xR2, which integrates IMU measurements as inputs and barometer and magnetometer measurements as outputs within a unified framework, guaranteeing local exponential stability (LES) under relaxed uniform observability conditions. The proposed approaches are validated using both simulated and real flight data. The results demonstrate that barometer-aided estimation provides a lightweight, reliable, and effective complementary sensing modality for attitude estimation in minimal-sensing configurations, offering a practical alternative when conventional velocity measurements are unavailable or degraded.
Chinese Translation
准确且稳健的姿态估计是自主车辆面临的一个关键挑战，特别是在全球导航卫星系统（GNSS）信号缺失的情况下以及在高度加速的飞行过程中。在这些条件下，单独依靠惯性测量单元（IMUs）进行可靠的倾斜估计是不够的，因为重力加速度和惯性加速度之间存在模糊性。尽管常用的辅助速度传感器如GNSS、皮托管、杜普勒雷达或视觉惯性里程计可能会被使用，但它们可能不可用、间歇性或成本高昂。本文提出了一种气压辅助的姿态估计架构，利用气压高度测量提供关于车辆垂直运动的补充信息，从而在SO(3)上的非线性观测器中增强姿态估计。贡献主要有两方面。首先，我们设计了一种确定性的Riccati观测器，与互补滤波器级联，确保在统一可观测性（UO）条件下几乎全局渐近稳定性（AGAS），同时保持姿态动力学的几何结构。其次，我们提出了一种在SO(3)xR2上演化的非线性观测器，该观测器将IMU测量作为输入，将气压计和磁力计测量作为输出，整合在一个统一框架内，保证在放宽的统一可观测性条件下的局部指数稳定性（LES）。所提出的方法通过模拟和真实飞行数据进行了验证。结果表明，气压辅助的估计为在最小传感配置下的姿态估计提供了一种轻量、可靠且有效的补充传感方式，为在传统速度测量不可用或退化时提供了一个实用的替代方案。
cs.RO / 10 / 2607.13312
Parsimonious disturbance-aware minimum-time planning with parametric uncertainty
考虑扰动的简约最小时间规划与参数不确定性
Abstract
This study presents and validates a minimum-lap-time planning (MLTP) framework for motorsport applications that embeds robustness against both state disturbances and parameter uncertainty. The methodology builds upon a prior disturbance-aware framework that, at each track point, propagates stochastic vehicle dynamics over a short horizon and tightens tyre-friction constraints based on the worst-case scenario at horizon end. We extend the formulation to account for uncertainty in key vehicle parameters: moment of inertia, centre-of-mass position, and aerodynamic drag coefficient. To keep the extended formulation computationally tractable, a spatially selective, parsimonious activation strategy confines the robust constraints to the circuit segments where they are most critical. We demonstrate the improved driveability of the robust references by employing a model predictive controller (MPC) as a virtual test driver. For each reference, the same MPC drives a simulated FSAE (Formula SAE) car over 1000 runs on a representative Barcelona-Catalunya sector, with randomly realised impulsive disturbances and parameter scatter. We compare a nominal reference, planned without robustness, against its robust counterparts. The latter yield consistently fewer failed runs and, at a moderate sector-time cost, show tighter dispersion of key signals (vehicle inputs, axle saturations) around the reference values, evidence of better trackability.
Chinese Translation
本研究提出并验证了一种用于赛车应用的最小圈时规划（MLTP）框架，该框架嵌入了对状态扰动和参数不确定性的鲁棒性。该方法建立在先前的扰动感知框架之上，在每个赛道点上，传播短时间范围内的随机车辆动力学，并根据时间范围结束时的最坏情况收紧轮胎摩擦约束。我们扩展了该公式，以考虑关键车辆参数的不确定性：惯性矩、质心位置和空气阻力系数。为了保持扩展公式的计算可行性，采用空间选择的简约激活策略，将鲁棒约束限制在最关键的赛道段。我们通过使用模型预测控制器（MPC）作为虚拟测试驾驶员，展示了鲁棒参考的改进可驾驶性。对于每个参考，MPC在代表性的巴塞罗那-加泰罗尼亚赛段上进行1000次模拟FSAE（Formula SAE）赛车的驾驶，随机实现冲击扰动和参数散布。我们将没有鲁棒性的名义参考与其鲁棒对应物进行比较。后者在失败运行次数上始终较少，并且在适度的赛段时间成本下，关键信号（车辆输入、车轴饱和度）围绕参考值的分散性更紧凑，显示出更好的可追踪性。
cs.RO / 11 / 2607.13319
Adapting Generalist Vehicle Models for High-Speed MPC Across Terrains
适应通用车辆模型以实现跨越不同地形的高速模型预测控制
Abstract
High-speed off-road autonomy requires precise closed-loop control for a target vehicle while remaining robust across changing terrains. Recent forward kinodynamic (FKD) prediction foundation models suggest a promising path, starting from a generalist model and specializing it to the target platform. However, effective specialization remains challenging, as it often requires substantial real-world data, and models adapted to one setting can still overfit to specific terrains or driving regimes. We present OptCar (Optimized Car), a recipe for bridging the gap from generalist to specialist FKD models that preserves cross-terrain generalization while optimizing performance for a specific vehicle. $\texttt{OptCar}$ introduces a history-conditioned dynamics adaptation module that encodes recent state-action observations into a dynamics context token, and then fine-tunes the generalist model using limited real-world data together with targeted synthetic rollouts from environment-specific system identification. In closed-loop model predictive control (MPC) experiments across three terrains and an out-of-distribution cart-pulling task, the largest gains appear at 6~m/s, the highest speed evaluated and the regime in which slip dominates tracking error. On vegetation and dirt, the most slip-diverse terrain, OptCar reduces 6~m/s trajectory tracking error by roughly 55% relative to a fine-tuned AnyCar baseline, and remains the most accurate even when an unseen cart payload changes the dynamics. With only 5 minutes of real data per terrain, OptCar is competitive on road with a specialist trained on 30 minutes of road data, and substantially outperforms it once the terrain changes.
Chinese Translation
高速越野自主驾驶需要对目标车辆进行精确的闭环控制，同时在变化的地形中保持稳健性。最近的前向运动动力学（FKD）预测基础模型提出了一条有前景的路径，从通用模型出发，专门化到目标平台。然而，有效的专门化仍然具有挑战性，因为这通常需要大量的真实世界数据，并且适应于一种设置的模型仍可能过拟合于特定地形或驾驶模式。我们提出了OptCar（优化车辆），这是一种弥合通用与专用FKD模型之间差距的方法，既保持跨地形的泛化能力，又优化特定车辆的性能。$ exttt{OptCar}$引入了一种基于历史的动态适应模块，该模块将最近的状态-动作观测编码为动态上下文标记，然后利用有限的真实世界数据和来自环境特定系统识别的目标合成滚动数据对通用模型进行微调。在三个地形和一个超出分布的推车任务的闭环模型预测控制（MPC）实验中，最大的增益出现在6~m/s，这是评估的最高速度，也是滑移主导跟踪误差的模式。在植被和泥土这两种滑移多样性最大的地形上，OptCar将6~m/s的轨迹跟踪误差相对于微调的AnyCar基线减少了约55%，即使在未见的推车负载改变动态的情况下，仍保持最高的准确性。仅用每个地形5分钟的真实数据，OptCar在道路上与经过30分钟道路数据训练的专用模型竞争，一旦地形发生变化，其表现则显著优于该专用模型。
cs.RO / 12 / 2607.13337
Design and Characterization of a Limb Encircling Actuator
肢体环绕驱动器的设计与表征
Abstract
Lower-limb powered exoskeletons have demonstrated substantial improvements in mobility and function, but most designs place actuation components lateral to the legs or remotely at the waist or back. These configurations often extend beyond the body's natural envelope, making devices intrusive in everyday use and potentially limiting societal adoption. We posit that rethinking actuation geometry could enable exoskeletons that conform more closely to the body. Here, we explored an actuation layout in which the actuator encircles the limb in a plane orthogonal to the limb axis, potentially reducing its spatial footprint around the body. We developed the Limb-Encircling Actuator (LEA) and characterized its electromechanical properties using a custom-built testbed. The LEA also features a novel radial bearing layout with potential as a lightweight or lower-cost alternative to traditional large-diameter bearings. The actuator achieved a continuous torque density of 7.5 Nm/kg with a mass of 894 g. Despite this high torque density and innovative layout, the system remained difficult to contain close to the body. These results highlight opportunities and challenges in limb-encircling actuation and provide insights into torque-dense exoskeleton designs that could integrate more readily into everyday apparel if challenges in actuator sizing and geometry are overcome.
Chinese Translation
下肢动力外骨骼在移动性和功能上表现出显著的改善，但大多数设计将驱动组件放置在腿部的侧面或远离身体的腰部或背部。这些配置往往超出了身体的自然轮廓，使得设备在日常使用中显得侵入性强，并可能限制社会的接受度。我们认为，重新思考驱动几何形状可能使外骨骼更贴合身体。在此，我们探讨了一种驱动布局，其中驱动器在垂直于肢体轴的平面上环绕肢体，可能减少其在身体周围的空间占用。我们开发了肢体环绕驱动器（Limb-Encircling Actuator, LEA），并使用定制测试平台表征其机电特性。LEA还具有新颖的径向轴承布局，可能成为传统大直径轴承的轻量化或低成本替代方案。该驱动器实现了7.5 Nm/kg的连续扭矩密度，质量为894克。尽管具有高扭矩密度和创新布局，该系统仍然难以紧贴身体。这些结果突显了肢体环绕驱动的机遇与挑战，并为扭矩密集型外骨骼设计提供了见解，如果能够克服驱动器尺寸和几何形状方面的挑战，这些设计可能更容易融入日常服装中。
cs.RO / 13 / 2607.13348
Safe Overtaking for Autonomous Racing Using Hierarchical Optimization and Learning-Based Control
基于层次优化和学习控制的自主赛车安全超车
Abstract
Autonomous racing overtaking requires balancing competitive performance with safety under nonlinear vehicle dynamics and real-time constraints. Model Predictive Control (MPC) combined with Control Barrier Functions (CBFs) provides a principled mechanism for certifying forward invariance of a safe set. However, commonly used fixed-decay discrete-time CBF formulations can become overly conservative in interactive racing scenarios, limiting overtaking performance and requiring manual tuning across track conditions. This paper proposes a hierarchical overtaking framework that explicitly separates maneuver-level decision making from safety-certified trajectory control, reducing conservatism while preserving safety. A high-level Mixed-Integer Quadratic Program (MIQP) resolves the combinatorial passing-side selection problem by selecting a feasible overtaking topology, while a nonlinear Frenet-frame MPC enforces vehicle dynamics and safety through embedded discrete-time CBF constraints. This decomposition isolates the combinatorial complexity of maneuver selection from the continuous trajectory optimization. To further mitigate the sensitivity of fixed-decay barrier constraints, a reinforcement learning policy adapts the discrete-time CBF decay parameter online, enabling context-dependent modulation of safety margins without directly controlling vehicle inputs. Simulation and scaled-hardware experiments show that no single fixed decay parameter achieves uniformly strong performance across tracks, whereas the adaptive strategy attains the highest aggregate success rate and consistently strong safety--performance trade-offs without per-track tuning, improving robustness to environment variation while maintaining safety constraint satisfaction in nominal operation.
Chinese Translation
自主赛车超车需要在非线性车辆动力学和实时约束下平衡竞争性能与安全性。模型预测控制（Model Predictive Control, MPC）结合控制屏障函数（Control Barrier Functions, CBFs）为验证安全集的前向不变性提供了一种原则性机制。然而，常用的固定衰减离散时间CBF公式在交互式赛车场景中可能变得过于保守，从而限制超车性能，并需要在不同赛道条件下进行手动调节。本文提出了一种层次化超车框架，明确将机动级决策与安全认证轨迹控制分开，减少保守性同时保持安全性。高层次的混合整数二次规划（Mixed-Integer Quadratic Program, MIQP）通过选择可行的超车拓扑来解决组合性超车侧选择问题，而非线性Frenet框架MPC则通过嵌入的离散时间CBF约束来强制执行车辆动力学和安全性。这种分解将机动选择的组合复杂性与连续轨迹优化隔离开来。为了进一步减轻固定衰减屏障约束的敏感性，强化学习策略在线调整离散时间CBF衰减参数，实现安全边际的上下文依赖调节，而无需直接控制车辆输入。仿真和缩放硬件实验表明，没有单一的固定衰减参数能够在各赛道上实现均匀强劲的性能，而自适应策略则在不需要逐赛道调节的情况下，获得了最高的整体成功率，并持续实现强大的安全性与性能权衡，提高了对环境变化的鲁棒性，同时在名义操作中保持了安全约束的满足。
cs.RO / 14 / 2607.13354
A Hybrid Sampling-Based Trajectory Planner with Game-Theoretic Guidance for Autonomous Racing
一种基于混合采样的轨迹规划器，结合博弈论指导用于自主赛车
Abstract
Autonomous racing demands planning algorithms that balance vehicle dynamics at the limits of handling with strategic decision-making in competitive multi-agent scenarios. Game theory provides a mathematical framework for modeling these interactions, enabling interactive trajectory planning and strategic behaviors, such as blocking. However, directly solving full dynamic games online is computationally prohibitive and challenging to integrate into robust, high-frequency autonomous software stacks. This paper proposes a hybrid architecture that integrates game-theoretic reasoning into a sampling-based motion planner, combining strategic interactions with robust trajectory generation. Building upon an $\alpha$-potential game formulation, we utilize an offline-learned potential function to capture multi-agent interactions. During online operation, a gradient-based optimization dynamically refines interaction parameters to generate an \textit{Interaction Reference Path}. This path serves as a dynamic cost bias within a high-frequency sampling planner. We evaluate our approach in a high-fidelity simulation environment on the Yas Marina Circuit. Qualitative and quantitative results demonstrate that our approach successfully induces defensive behaviors like blocking without carrying the computational burden of full dynamic game solvers.
Chinese Translation
自主赛车要求规划算法在车辆动态与竞争多智能体场景中的战略决策之间取得平衡。博弈论提供了一种数学框架，用于建模这些交互，支持交互式轨迹规划和战略行为，如阻挡。然而，在线直接求解完整动态博弈在计算上是不可行的，并且难以集成到稳健的高频自主软件堆栈中。本文提出了一种混合架构，将博弈论推理集成到基于采样的运动规划器中，结合战略交互与稳健的轨迹生成。在$eta$-潜力博弈的基础上，我们利用离线学习的潜力函数来捕捉多智能体交互。在在线操作期间，基于梯度的优化动态地细化交互参数，以生成 extit{交互参考路径}。该路径作为高频采样规划器中的动态成本偏置。我们在雅斯玛丽娜赛道的高保真仿真环境中评估了我们的方法。定性和定量结果表明，我们的方法成功地诱导了如阻挡等防御行为，而无需承担完整动态博弈求解器的计算负担。
cs.RO / 15 / 2607.13403
Min-Max Regret Task Allocation and Planning of Heterogeneous Multi-Robot System in Partially Known Environments
部分已知环境中异构多机器人系统的最小-最大遗憾任务分配与规划
Abstract
Efficient task allocation for large-scale Heterogeneous Multi-Robot Systems (HMRS) is critical, yet dealing with complex temporal logic tasks in partially known environment (PKE) remains a computational bottleneck. Existing approaches often struggle to balance exploring uncertain regions and exploiting known resources, while also suffering from exponential computational complexity. To address these issues, this paper presents a robust planning framework that simultaneously handles high-level logical constraints and environmental uncertainty without sacrificing scalability. We formulate the problem as a min-max regret optimization, proposing a Region-Binding Atomic Proposition (RbAP) to capture resource uncertainty within the automaton structure. To solve this, we propose the Extended Planning Decision Tree (E-PDT) equipped with a novel Regret-based Branch-and-Bound (BnB) strategy. Unlike traditional methods that rely on prior probabilities or worst-case analysis, our approach dynamically prunes suboptimal policies, effectively balancing the need for information gathering (exploration) and task completion (exploitation). Theoretical analysis confirms the feasibility and completeness of our approach. Extensive numerical and physical experiments demonstrate that the proposed framework achieves near-linear scalability with respect to the number of robots and types, significantly outperforming MILP-based baselines in both solution quality and computational efficiency.
Chinese Translation
高效的任务分配对于大规模异构多机器人系统（HMRS）至关重要，但在部分已知环境（PKE）中处理复杂的时序逻辑任务仍然是一个计算瓶颈。现有方法往往难以平衡对不确定区域的探索与对已知资源的利用，同时也面临指数级的计算复杂性。为了解决这些问题，本文提出了一种稳健的规划框架，能够同时处理高层次的逻辑约束和环境不确定性，而不牺牲可扩展性。我们将问题形式化为最小-最大遗憾优化，提出了一种区域绑定原子命题（RbAP）来捕捉自动机结构中的资源不确定性。为了解决这个问题，我们提出了扩展规划决策树（E-PDT），并配备了一种新颖的基于遗憾的分支限界（BnB）策略。与依赖于先验概率或最坏情况分析的传统方法不同，我们的方法动态修剪次优策略，有效平衡信息收集（探索）和任务完成（利用）的需求。理论分析确认了我们方法的可行性和完整性。大量的数值和物理实验表明，所提出的框架在机器人数量和类型方面实现了近线性的可扩展性，在解决方案质量和计算效率上显著优于基于MILP的基线。
cs.RO / 16 / 2607.13405
WNOJ-LIO: A White-Noise-on-Jerk Motion-Prior EKF for High-Dynamic LiDAR-IMU Fusion
WNOJ-LIO：一种基于白噪声-加速度变化运动先验的高动态激光雷达-惯性测量单元融合扩展卡尔曼滤波器
Abstract
LiDAR-inertial odometry (LIO) is a key component of autonomous navigation, but high-dynamic driving exposes two coupled challenges: intra-scan motion distortion and vibration-contaminated inertial measurements. Most real-time LiDAR-inertial pipelines propagate the system state by integrating raw IMU measurements and then use the propagated trajectory for point cloud de-distortion, thereby propagating inertial noise into both the corrected scan and the subsequent scan-to-map registration. This paper presents WNOJ-LIO, a LiDAR-IMU fusion framework based on a White-Noise-on-Jerk (WNOJ) Extended Kalman Filter (EKF). WNOJ-LIO employs a decoupled WNOJ prior on $\R^3 \times \SO(3)$ for state prediction and treats the IMU as a high-frequency measurement source rather than the driver of state propagation. The resulting posterior state history is then used for LiDAR scan de-distortion and subsequent point-to-plane LiDAR updates. The decoupled process model enables closed-form covariance propagation, thereby bridging the gap between batch WNOJ Gaussian process (GP) trajectory priors and recursive filtering. Simulation results demonstrate improvements in acceleration and angular-velocity denoising, scan de-distortion, and localization accuracy over a FAST-LIO-style baseline. Real-world experiments were conducted using an autonomous racing car on four driving segments with maximum speeds ranging from 53 to 208~km/h, covering a wide range of vehicle vibration levels. The experiments further validate the proposed method and provide a comprehensive evaluation of its performance in estimating acceleration, angular velocity, body-frame linear velocity, attitude, and position under highly dynamic driving. The source code of WNOJ-LIO is publicly available at https://github.com/LvJohny/wnoj-ekf-lio.git.
Chinese Translation
激光雷达惯性里程计（LIO）是自主导航的关键组成部分，但高动态驾驶带来了两个相互关联的挑战：扫描内运动失真和振动污染的惯性测量。大多数实时激光雷达-惯性管道通过整合原始惯性测量单元（IMU）测量值来传播系统状态，然后使用传播的轨迹进行点云去失真，从而将惯性噪声传播到校正后的扫描和后续的扫描与地图配准中。本文提出了WNOJ-LIO，一种基于白噪声-加速度变化（WNOJ）扩展卡尔曼滤波器（EKF）的激光雷达-IMU融合框架。WNOJ-LIO在$ ext{R}^3 imes ext{SO}(3)$上采用解耦的WNOJ先验进行状态预测，并将IMU视为高频测量源，而非状态传播的驱动因素。由此得到的后验状态历史用于激光雷达扫描去失真和后续的点对平面激光雷达更新。解耦的过程模型实现了闭式形式的协方差传播，从而弥合了批处理WNOJ高斯过程（GP）轨迹先验与递归滤波之间的差距。仿真结果表明，与FAST-LIO风格基线相比，在加速度和角速度去噪、扫描去失真和定位精度方面都有所改善。通过使用一辆自主赛车在四个驾驶段进行的实地实验，最大速度范围从53到208 km/h，覆盖了广泛的车辆振动水平。这些实验进一步验证了所提出的方法，并全面评估了其在高动态驾驶下估计加速度、角速度、机体坐标系线速度、姿态和位置的性能。WNOJ-LIO的源代码已公开，地址为https://github.com/LvJohny/wnoj-ekf-lio.git。
cs.RO / 17 / 2607.13410
Ego-Dynamics-Augmented World Model for Autonomous Driving with Zero-Shot Cross-Chassis Adaptation
增强自我动态的世界模型用于零样本跨底盘适应的自主驾驶
Abstract
World model (WM)-based reinforcement learning enables sample-efficient end-to-end autonomous driving learning by imagining long-horizon trajectories in latent space. However, most driving WMs operate on bird's-eye-view (BEV) representations that are inherently egocentric: the transition between consecutive frames entangles the ego vehicle's own motion with scene dynamics. As a result, the WM devotes significant capacity to recovering ego-motion from warped observations, at the cost of scene modeling fidelity and imagination accuracy. This work proposes DynaDreamer, a dynamics-augmented Dreamer-style reinforcement learning method to address this problem by augmenting the WM with an explicit ego-dynamics prior. A physics-informed ego-dynamics encoder-decoder extracts the ego-state history into a compact and identifiable context, which modulates a causal Transformer WM to condition both its prior and posterior latents. During imagination, the ego-dynamics predictor propagates this context forward to keep the ego-dynamics prior synchronized with the rollout. An information-theoretic analysis shows that conditioning on this context reduces both the predictive entropy of the observation transition and the prior--posterior Kullback--Leibler divergence, confining the WM's modeling burden to the scene dynamics beyond ego-motion. An additional benefit is zero-shot cross-chassis adaptation: the ego-dynamics context depends on identifiable chassis parameters, so that a vehicle with previously unseen dynamic characteristics can adapt the WM to the new chassis without retraining. Experiments demonstrate that DynaDreamer improves task success rates over the strongest baseline by 28% and 61% in urban and highway driving scenarios, respectively, with the advantage rising to 73% when extrapolating to unseen chassis.
Chinese Translation
基于世界模型（WM）的强化学习通过在潜在空间中想象长时间轨迹，实现了样本高效的端到端自主驾驶学习。然而，大多数驾驶WM操作于鸟瞰图（BEV）表示，这种表示本质上是自我中心的：连续帧之间的过渡将自我车辆的运动与场景动态纠缠在一起。因此，WM在从扭曲的观测中恢复自我运动方面投入了大量能力，牺牲了场景建模的保真度和想象的准确性。本研究提出了DynaDreamer，一种增强动态的Dreamer风格强化学习方法，通过用显式的自我动态先验增强WM来解决这一问题。一个物理信息驱动的自我动态编码器-解码器将自我状态历史提取为紧凑且可识别的上下文，调节因果Transformer WM以条件其先验和后验潜变量。在想象过程中，自我动态预测器向前传播该上下文，以保持自我动态先验与展开同步。信息论分析表明，基于该上下文的条件减少了观测过渡的预测熵和先验-后验Kullback-Leibler散度，将WM的建模负担限制在超出自我运动的场景动态上。另一个好处是零样本跨底盘适应：自我动态上下文依赖于可识别的底盘参数，因此具有先前未见动态特征的车辆可以在不重新训练的情况下将WM适应于新底盘。实验表明，DynaDreamer在城市和高速公路驾驶场景中，任务成功率分别比最强基线提高了28%和61%，当推广到未见底盘时，优势上升至73%。
cs.RO / 18 / 2607.13429
Generalizable VLA Finetuning via Representation Anchoring and Language-Action Alignment
通过表示锚定和语言-动作对齐实现可泛化的视觉-语言-动作微调
Abstract
Finetuning a pretrained vision-language model (VLM) on robot demonstrations via behavior cloning (BC) has become the standard recipe for vision-language-action (VLA) policies. However, BC finetuning progressively overwrites the pretrained representations that support visual and semantic generalization. Co-training on web image-text data, a common remedy, does not prevent this; it applies language and action losses to separate observations, leaving VLAs with language-action misalignment that standard manipulation benchmarks do not expose. We propose Anchor-Align, which augments BC with two objectives: Vision-Language Anchoring distills layer-wise representations from a frozen VLM copy to prevent this drift, while Language-Action Alignment converts each action target into a discrete motion-direction label and jointly trains language and action prediction on the same robot observation. On a physical xArm7 robot, across two widely used VLA architectures, Anchor-Align improves real-robot success on both (28% to 54% and 37% to 60%). At scale in simulation, we demonstrate consistent improvements on OOD perturbations, perceptual robustness, and long-horizon control across LIBERO-PRO, LIBERO-Plus, and CALVIN, respectively, suggesting that preserving pretrained representations and effective action learning are not fundamentally at odds. Project page: anchoralignvla.github.io
Chinese Translation
通过行为克隆（BC）对预训练的视觉-语言模型（VLM）进行微调，已成为视觉-语言-动作（VLA）策略的标准方法。然而，BC微调逐渐覆盖了支持视觉和语义泛化的预训练表示。共同训练网络图像-文本数据是一种常见的补救措施，但并不能阻止这一现象；它将语言和动作损失应用于不同的观测，导致VLA在语言-动作上存在不一致，而标准的操作基准并未暴露这一问题。我们提出了Anchor-Align，它通过两个目标增强BC：视觉-语言锚定从一个冻结的VLM副本中提取逐层表示，以防止这种漂移，而语言-动作对齐则将每个动作目标转换为离散的运动方向标签，并在同一机器人观测上联合训练语言和动作预测。在物理xArm7机器人上，针对两种广泛使用的VLA架构，Anchor-Align在真实机器人成功率上均有所提升（从28%提高到54%，从37%提高到60%）。在模拟环境中，我们在LIBERO-PRO、LIBERO-Plus和CALVIN上展示了一致的改进，涵盖了OOD扰动、感知鲁棒性和长时间控制，这表明保持预训练表示和有效的动作学习并不根本矛盾。项目页面：anchoralignvla.github.io
cs.RO / 19 / 2607.13444
Stress-Sharing: A Bio-Inspired Approach to Decentralized Fault Repair in Modular Spacecraft
应力共享：一种生物启发的模块化航天器去中心化故障修复方法
Abstract
Structural damage in modular spacecraft can disrupt mechanical and communication connectivity, reducing system capability. Existing approaches rely on redundancy or preplanned reconfiguration and do not enable autonomous repair under local information and physical constraints. We model the spacecraft as a lattice-constrained graph and introduce a fully decentralized, asynchronous stress-sharing repair policy inspired by biological wound healing: local distress signals guide surviving modules toward damaged regions to close fragmented gaps, after which each displaced module locally retraces its own motions to recover the pre-damage shape, using only local information and no absolute position sensing. We evaluate the policy in PyBullet rigid-body simulation across structures of up to 160 modules, three fault densities (10, 20, 30%), and random and localized damage. The policy consolidates the surviving modules into a single connected body: even in the most severe case tested, where 30% of modules fail at random, it gathers roughly 80% or more of the surviving modules into one connected component, and this fraction improves with assembly size, making the approach well suited as a swarm-scale repair policy for large modular spacecraft.
Chinese Translation
模块化航天器中的结构损伤会 disrupt 机械和通信连接，降低系统能力。现有的方法依赖于冗余或预先规划的重配置，并不支持在局部信息和物理约束下的自主修复。我们将航天器建模为一个晶格约束图，并引入一种完全去中心化的、异步的应力共享修复策略，该策略受到生物伤口愈合的启发：局部 distress 信号引导存活模块朝向损坏区域，以关闭破碎的间隙，随后每个位移模块在仅使用局部信息且不依赖绝对位置感知的情况下，局部回溯其自身的运动以恢复到损伤前的形状。我们在 PyBullet 刚体仿真中评估该策略，涉及多达 160 个模块的结构、三种故障密度（10%、20%、30%）以及随机和局部损伤。在最严重的测试案例中，即 30% 的模块随机故障，该策略将大约 80% 或更多的存活模块聚集成一个连接组件，并且这一比例随着组装规模的增加而改善，使得该方法非常适合作为大规模模块化航天器的群体修复策略。
cs.RO / 20 / 2607.13451
Learning Physics-Guided Residual Dynamics for Deformable Object Simulation
学习物理引导的残差动态以进行可变形物体模拟
Abstract
Simulating deformable objects is essential for a wide range of robotic manipulation applications, yet accurately predicting their dynamics remains challenging. We propose Physics-Guided Residual Dynamics (PGRD), a hybrid simulation framework that combines the advantages of physics-based and learning-based approaches. Specifically, PGRD combines an optimizable spring-mass simulator as a backbone with a learned neural network that predicts residual corrections to the physics-based predictions. We adopt a velocity-based formulation to ensure stable simulation and a sliding-window transformer architecture to capture temporal dependencies. We show that PGRD produces more accurate results than both purely physics-based and learning-based methods on a set of diverse real-world deformable objects. We further demonstrate the utility of PGRD in two applications: manipulation planning via Model Predictive Control, including a language-conditioned setting with a generated goal image; and interactive simulation via action-conditioned video prediction by 3D Gaussian Splatting.
Chinese Translation
模拟可变形物体对于广泛的机器人操控应用至关重要，但准确预测其动态仍然具有挑战性。我们提出了一种物理引导的残差动态（Physics-Guided Residual Dynamics, PGRD），这是一种混合模拟框架，结合了基于物理和基于学习的方法的优点。具体而言，PGRD将一个可优化的弹簧-质量模拟器作为基础，并与一个学习的神经网络相结合，该网络预测对基于物理的预测的残差修正。我们采用基于速度的公式以确保稳定的模拟，并使用滑动窗口变换器架构来捕捉时间依赖性。我们展示了PGRD在一组多样的真实世界可变形物体上产生的结果比纯粹的基于物理和基于学习的方法更为准确。我们进一步展示了PGRD在两个应用中的实用性：通过模型预测控制进行操控规划，包括一个带有生成目标图像的语言条件设置；以及通过动作条件的视频预测进行交互式模拟，采用3D高斯点云技术。
cs.RO / 21 / 2607.13455
Reverse to Advance: Teleoperation-Cost Effective Hard Policy Learning from Reversed Easy Tasks
反向推进：从反向简单任务中进行成本有效的远程操作硬策略学习
Abstract
High-quality teleoperation datasets are costly to collect, particularly for hard tasks. We observe that many tasks exhibit directional asymmetry: completing the forward hard task is difficult, whereas reversing it by relaxing or disrupting the environment is comparatively easy. This suggests that reversed easy-task trajectories can serve as a scalable supervision signal for the hard task, reducing the cost of manual demonstration collection. However, reversed data can be noisy, and directly training on it may yield suboptimal policies. To enable largely automated acquisition and effective use of reversed data, we propose a teleoperation-cost effective framework for hard policy learning via temporal reversal of easy tasks, consisting of three key components: a closed-loop data collection pipeline that alternates between hard-task and easy-task policies to autonomously reset the environment and generate diverse trajectories; a hierarchical data refinement pipeline that temporally inverts easy-task rollouts and filters low-quality motion using kinematic priors and a critic-guided advantage filter; and an iterative policy learning method that trains the hard-task policy using both initial reversed easy-task demonstrations and the filtered reversed data in a continuous online learning loop. By combining automated collection, hierarchical refinement, and iterative learning, our method enables scalable, reliable training of complex, high-precision manipulation tasks. Across two simulated benchmarks and real-robot experiments, we demonstrate that our method improves hard-task success rates with higher data efficiency and more stable training compared to reversal-based and reinforcement-learning baselines, without requiring extensive hard-task teleoperation.
Chinese Translation
高质量的远程操作数据集收集成本高昂，尤其是在处理困难任务时。我们观察到许多任务表现出方向性不对称：完成前向的困难任务很难，而通过放松或干扰环境反向执行则相对容易。这表明，反向简单任务的轨迹可以作为困难任务的可扩展监督信号，从而降低人工演示收集的成本。然而，反向数据可能会噪声较大，直接在其上进行训练可能会导致次优策略。为了实现大规模自动化获取和有效利用反向数据，我们提出了一种通过简单任务的时间反转进行成本有效的远程操作硬策略学习框架，包含三个关键组件：一个闭环数据收集管道，该管道在困难任务和简单任务策略之间交替，以自主重置环境并生成多样化轨迹；一个分层数据精炼管道，该管道对简单任务的回放进行时间反转，并使用运动学先验和批评引导的优势过滤器过滤低质量运动；以及一种迭代策略学习方法，该方法使用初始反向简单任务演示和过滤后的反向数据在连续在线学习循环中训练困难任务策略。通过结合自动化收集、分层精炼和迭代学习，我们的方法实现了复杂高精度操作任务的可扩展、可靠训练。在两个模拟基准和真实机器人实验中，我们证明了我们的方法在数据效率和训练稳定性方面优于基于反转和强化学习的基线，同时无需大量的困难任务远程操作。
cs.RO / 22 / 2607.13461
Joint On-and-Off Policy Learning for Vision-and-Language Navigation
视觉与语言导航的联合在线与离线策略学习
Abstract
Vision-and-Language Navigation (VLN) necessitates an embodied agent to navigate in the physical world by adhering to natural language instructions. Recent advancements in Vision-Language Models (VLM) have propelled the development of VLM-based VLN methods with two predominant paradigms: (1) imitation learning (IL) on expert demonstrations, followed by the Dataset Aggregation (DAgger) algorithm to bolster error recovery capabilities; (2) reinforcement learning (RL) driven by verifiable rewards to enhance reasoning and exploration. A notable gap is the absence of integration between these two distinct paradigms. This paper introduces JOP-VLN, a novel VLN framework that synergistically combines off-policy imitation learning and on-policy exploration within a three-stage training pipeline. Initially, IL is employed on expert demonstrations to acquire basic navigation skills. Subsequently, the DAgger algorithm is utilized to generate heuristic exploration trajectories, which are then used for imitation learning to improve error recovery capabilities. Finally, a joint on-and-off policy learning framework is implemented, featuring high-entropy trajectory sampling to enhance RL training efficiency and an error-correction-prioritized trajectory sorting strategy for effective error correction. Extensive experiments demonstrate the efficacy of JOP-VLN, achieving success rates of 69.9% and 68.0% on the VLN-CE R2R and RxR benchmarks, respectively, setting a new state-of-the-art on R2R. Project page: https://qingrongh.github.io/JOP-VLN.
Chinese Translation
视觉与语言导航（VLN）要求具身代理根据自然语言指令在物理世界中进行导航。最近，视觉语言模型（VLM）的进展推动了基于VLM的VLN方法的发展，主要有两种范式：（1）在专家示范上进行模仿学习（IL），随后使用数据集聚合（DAgger）算法增强错误恢复能力；（2）通过可验证奖励驱动的强化学习（RL），以提高推理和探索能力。一个显著的缺口是这两种不同范式之间缺乏整合。本文提出了JOP-VLN，一个新颖的VLN框架，协同结合了离线模仿学习和在线探索，采用三阶段训练流程。首先，在专家示范上应用IL以获取基本导航技能。随后，利用DAgger算法生成启发式探索轨迹，这些轨迹用于模仿学习以改善错误恢复能力。最后，实施了一个联合在线与离线策略学习框架，具有高熵轨迹采样以提高RL训练效率，以及一种以错误修正为优先的轨迹排序策略，以实现有效的错误修正。大量实验表明JOP-VLN的有效性，在VLN-CE R2R和RxR基准上分别达到了69.9%和68.0%的成功率，在R2R上设定了新的最先进水平。项目页面：https://qingrongh.github.io/JOP-VLN。
cs.RO / 23 / 2607.13472
EgoHTR: Egocentric 4D Demonstrations of Human Terrain Traversal
EgoHTR：以自我为中心的人类地形穿越的四维演示
Abstract
Deploying humanoid robots in unstructured terrain remains an open problem. While classic reinforcement learning struggles with the sheer complexity of real-world interactions, more promising methods leveraging human priors remain limited to models lacking contextual awareness. The restricted motion synthesis is a direct consequence of existing dataset pipelines failing to capture human-scene sequences in challenging environments. To bridge this gap between humanoid learning and scene reconstruction, we introduce the Egocentric Human-Terrain Reconstruction (EgoHTR) dataset. We develop and open-source a reconstruction pipeline capturing 55 scene-aligned 4D human motion sequences in diverse, complex environments using a multi-sensor setup of egocentric wearables and a portable 3D scanner. The resulting dataset comprises over 150k frames, which we evaluate against motion-capture ground truth, demonstrating state-of-the-art accuracy and establishing a rigorous benchmark for human motion analysis and synthesis. Further, we leverage this data to train perceptive locomotion policies, demonstrating hardware deployment on a Unitree G1 for reconstructed reference motions. Our pipeline enables community-driven dataset extensions and factors the problem to help researchers build foundational, context-aware robots that reliably traverse uneven terrain.
Chinese Translation
在非结构化地形中部署类人机器人仍然是一个未解决的问题。尽管经典的强化学习在处理真实世界交互的复杂性方面面临挑战，但利用人类先验知识的更有前景的方法仍然局限于缺乏上下文意识的模型。现有数据集管道未能捕捉到在复杂环境中人类与场景的序列，导致运动合成受到限制。为了弥补类人学习与场景重建之间的差距，我们引入了以自我为中心的人类-地形重建（EgoHTR）数据集。我们开发并开源了一个重建管道，使用以自我为中心的可穿戴设备和便携式3D扫描仪在多样且复杂的环境中捕捉到55个场景对齐的四维人类运动序列。生成的数据集包含超过15万帧，我们将其与运动捕捉的真实数据进行评估，展示了最先进的准确性，并为人类运动分析和合成建立了严格的基准。此外，我们利用这些数据训练感知性步态策略，并在Unitree G1上展示了重建参考运动的硬件部署。我们的管道支持社区驱动的数据集扩展，并将问题分解，帮助研究人员构建基础的、具备上下文意识的机器人，以可靠地穿越不平坦的地形。
cs.RO / 24 / 2607.13475
Deformable State Estimation for Autonomous Surgical Tissue Retraction Under Partial Observability
在部分可观测条件下用于自主外科组织回缩的可变形状态估计
Abstract
Surgical tissue retraction requires effective manipulation planning under partial and noisy perception. We study state estimation for deformable tissue retraction, where only sparse observations of the tissue surface are available at decision time. We propose a learned state estimator that reconstructs the full deformable mesh state from 40 noisy vertex observations. The estimator combines a multilayer perceptron with a low-dimensional PCA latent representation and is trained using geometry-aware regularization that encourages smooth and physically plausible deformations. We evaluate the approach in a 2D deformable sheet simulation using single-step and multi-step retraction planning. Results show that the learned estimator achieves 98.1% of oracle performance in multi-step retraction while supporting efficient inference. These results demonstrate that learned, geometry-regularized state estimation can support effective deformable manipulation under realistic perception constraints.
Chinese Translation
外科组织回缩需要在部分和噪声感知下进行有效的操作规划。我们研究了可变形组织回缩的状态估计，其中在决策时仅可获得稀疏的组织表面观察。我们提出了一种学习的状态估计器，该估计器从40个噪声顶点观察中重建完整的可变形网格状态。该估计器结合了多层感知器与低维主成分分析（PCA）潜在表示，并使用几何感知的正则化进行训练，以鼓励平滑和物理上合理的变形。我们在一个二维可变形薄片模拟中评估了该方法，使用单步和多步回缩规划。结果表明，在多步回缩中，学习的估计器达到了98.1%的oracle性能，同时支持高效推理。这些结果表明，学习的几何正则化状态估计可以在现实感知约束下支持有效的可变形操作。
cs.RO / 25 / 2607.13478
VAMP-MR: Vector-Accelerated Motion Planning and Execution for Multi-Robot-Arms
VAMP-MR：面向多机器人臂的向量加速运动规划与执行
Abstract
Multi-robot-arm motion planning is a key challenge in deploying multiple manipulators for industrial tasks such as manufacturing. Existing search-based and sampling-based solvers often require significant computation time to produce collision-free, high-quality motions suitable for safe real-world execution. In this work, we introduce a new suite of multi-robot-arm motion planners capable of near real-time motion generation, combining classical planning algorithms with state-of-the-art vectorized collision-checking techniques. Based on CPU SIMD instructions, our new planners accelerate their primary bottleneck, collision checking, and achieve up to two orders of magnitude speedup in both motion planning and execution postprocessing for multi-arm manipulation tasks. We also release our implementation to lower the barrier for research and development of multi-robot-arm planning and manipulation problems. Code is available at https://vamp-mr.github.io/vamp-mr
Chinese Translation
多机器人臂的运动规划是部署多个操纵器进行工业任务（如制造）中的一个关键挑战。现有的基于搜索和基于采样的求解器通常需要显著的计算时间来生成适合安全实际执行的无碰撞高质量运动。在本研究中，我们介绍了一套新的多机器人臂运动规划器，能够实现近实时的运动生成，结合了经典规划算法和最先进的向量化碰撞检测技术。基于CPU SIMD指令，我们的新规划器加速了其主要瓶颈——碰撞检测，并在多臂操纵任务的运动规划和执行后处理上实现了高达两个数量级的加速。我们还发布了我们的实现，以降低多机器人臂规划和操纵问题的研究与开发门槛。代码可在 https://vamp-mr.github.io/vamp-mr 获取。
cs.RO / 26 / 2607.13479
Topology-Agnostic Mesh Reconstruction of Deformable Objects from Sparse Touch
无拓扑依赖的可变形物体稀疏触摸网格重建
Abstract
Estimating the full shape of a deformable object is especially challenging when vision is unavailable: in the dark, inside an opaque bag, behind the manipulating hand, or under heavy self-occlusion. Touch is the natural sensor in these settings, but touches are sparse and local. We present a single topology-agnostic estimator that reconstructs the full mesh of a deformable object from only a few touches and no vision, using one permutation-invariant cross-attention architecture that handles a 1D rope, a 2D cloth, and a 3D volumetric soft body. The learned estimator reduces reconstruction error by roughly two-thirds relative to non-learned geometric mesh completion and a Gaussian-process surface baseline, and it outperforms a simpler global-pool set encoder, with the gap growing as more touches are observed. We then show that the estimator's deep-ensemble uncertainty can be used to learn where to touch next, which lowers error further and beats both random touching and a Gaussian-process active baseline at sparse budgets. This gain is modest on average but grows with self-occlusion and on the error tail. When vision is also available, where to touch barely matters, motivating the vision-free setting we study.
Chinese Translation
在缺乏视觉信息的情况下，估计可变形物体的完整形状尤其具有挑战性：在黑暗中、在不透明的袋子里、在操控手的后面或在严重自遮挡的情况下。触摸是这些环境中的自然传感器，但触摸是稀疏和局部的。我们提出了一种单一的无拓扑依赖估计器，该估计器仅通过少量触摸和无视觉信息重建可变形物体的完整网格，使用一种排列不变的交叉注意力架构，能够处理一维绳索、二维布料和三维体积软体。所学习的估计器相较于非学习的几何网格补全和高斯过程表面基线，减少了大约三分之二的重建误差，并且在观察到更多触摸时，其性能优于更简单的全局池集编码器。我们进一步展示了估计器的深度集成不确定性可以用于学习下一步触摸的位置，这进一步降低了误差，并在稀疏预算下优于随机触摸和高斯过程主动基线。尽管这一增益在平均水平上是适度的，但在自遮挡和误差尾部时则有所增加。当视觉信息可用时，触摸的位置几乎没有影响，这进一步激励了我们所研究的无视觉设置。
cs.RO / 27 / 2607.13497
Layered Risk Mapping for Autonomous Patient Transport in Expeditionary Medical Facilities
用于远征医疗设施中自主患者运输的分层风险映射
Abstract
In expeditionary medical facilities, routine patient transport imposes a compounding burden of personal protective equipment consumption, staff diversion, and elevated infection risk that becomes unsustainable under surge conditions. While autonomous wheelchairs could absorb this operational load, the safety-critical nature of patient transit within these highly unstructured and dynamic environments poses complex navigational challenges. To address this, we present a layered risk mapping framework that fuses four heterogeneous environmental hazards (terrain slope, static and dynamic obstacles, and semantic traversability) into a unified probabilistic cost surface via a Noisy-OR fusion model. In a paired Monte-Carlo evaluation, risk-informed fusion reduces collision rates from over 73% to under 32% and more than doubles obstacle clearance relative to a risk-unaware baseline. Additionaly, Noisy-OR achieves the highest clearance to obstacles and the lowest conditional peak risk across all tested hazard densities. We further validate the framework on a commercial powered wheelchair across three representative mission profiles in indoor and outdoor deployments, demonstrating that this architecture successfully meets the planning requirements of this previously unaddressed operational regime.
Chinese Translation
在远征医疗设施中，常规患者运输带来了个人防护装备消耗、人员分流和感染风险增加的复合负担，这在突发情况下变得不可持续。虽然自主轮椅可以吸收这一操作负担，但在这些高度非结构化和动态环境中，患者运输的安全关键性质带来了复杂的导航挑战。为了解决这一问题，我们提出了一种分层风险映射框架，该框架通过Noisy-OR融合模型将四种异构环境危害（地形坡度、静态和动态障碍物以及语义可通行性）融合为统一的概率成本表面。在配对的蒙特卡洛评估中，基于风险的信息融合将碰撞率从超过73%降低到低于32%，并且相对于风险无知的基线，障碍物清除率增加了两倍以上。此外，Noisy-OR在所有测试的危害密度中实现了最高的障碍物清除率和最低的条件峰值风险。我们进一步在商业电动轮椅上验证了该框架，在室内和室外部署的三个代表性任务配置中，证明该架构成功满足了这一先前未解决的操作领域的规划要求。
cs.RO / 28 / 2607.13516
Improving Map Consistency in Graph-Based LiDAR SLAM Through Information-Aware Odometry and Retroactive Loop Closure
通过信息感知里程计和回溯闭环来提高基于图的激光雷达SLAM中的地图一致性
Abstract
High-quality maps are fundamental for robotics tasks such as navigation and planning. Although modern graph-based LiDAR SLAM systems achieve good trajectory accuracies, a low trajectory error alone does not guarantee geometrically consistent maps, particularly at revisit locations where missed loop closures and residual drift can produce local misalignments. In this work, we address the problem of jointly improving global trajectory estimation and local map quality in 3D LiDAR SLAM. We first propose a framework to efficiently estimate geometry-dependent information matrices for ICP, enabling principled weighting of odometry constraints in a pose graph. We then introduce a hierarchical loop-closure module that decouples place recognition from geometric registration, together with a retroactive loop-closure module that exploits the optimized pose graph to recover missed loop closures. We also propose an evaluation protocol to measure map consistency at revisit locations. We evaluate our SLAM system on several datasets against state-of-the-art LiDAR SLAM systems. Experimental results demonstrate global trajectory accuracies on par with or better than existing methods while consistently improving local geometric map consistency at revisit locations. These results suggest that coupling uncertainty-aware odometry with geometry-guided loop-closure refinement leads to more accurate trajectories and higher-quality maps.
Chinese Translation
高质量地图是导航和规划等机器人任务的基础。尽管现代基于图的激光雷达SLAM系统在轨迹精度方面表现良好，但仅仅低轨迹误差并不能保证几何一致的地图，特别是在重访位置，漏掉的闭环和残余漂移可能导致局部错位。在本研究中，我们解决了在3D激光雷达SLAM中共同改善全局轨迹估计和局部地图质量的问题。我们首先提出了一个框架，以高效估计与几何相关的信息矩阵，用于ICP，从而在姿态图中实现里程计约束的原则性加权。然后，我们引入了一个层次化的闭环模块，将地点识别与几何配准解耦，同时引入一个回溯闭环模块，利用优化后的姿态图来恢复漏掉的闭环。我们还提出了一种评估协议，用于测量重访位置的地图一致性。我们在多个数据集上对我们的SLAM系统进行了评估，并与最先进的激光雷达SLAM系统进行了比较。实验结果表明，我们的全局轨迹精度与现有方法相当或更好，同时在重访位置持续改善局部几何地图一致性。这些结果表明，将不确定性感知的里程计与几何引导的闭环精炼相结合，可以获得更准确的轨迹和更高质量的地图。
cs.RO / 29 / 2607.13522
Kepler-Encoder-v0.1: Towards a Multimodal Embedding Model for Robots
Kepler-Encoder-v0.1：面向机器人的多模态嵌入模型
Abstract
A robot must understand the state of its own body, but a camera sees only part of it. Force and contact leave almost no trace in a single frame, and raw vision features read force at $R^2$ at or below $0.10$ on every robot we test. We present Kepler-Encoder-v0.1, a robot-first multimodal encoder that treats robot state as a modality and fuses vision, proprioception, and force/torque into a single shared latent with a learned-query cross-attention layer, trained self-supervised by masked cross-modal prediction under the LeJEPA/SIGReg objective. At evaluation only vision enters, which poses a sharp question. Does fusing state into training make the vision-only latent carry anything the pixels do not already contain? On the RH20T corpus the answer is yes, precisely where the camera is weakest. On held-out scenes, the vision-only latent recovers end-effector state, and force in particular, significantly above both raw frozen-ViT features and a compute-matched vision-only control on every sensored robot, though absolute force recovery at a single timestep is modest; on motor state, which the camera largely sees, it is statistically tied with the strongest vision baselines, and it is the only feature whose latent geometry tracks state. A single embodiment-agnostic encoder covers four robots, and a data-matched control shows this breadth reflects embodiment diversity rather than data volume. The frozen latent is directly useful. Its own cross-modal prediction error is a training-free invalid-state monitor (AUROC $0.90$ on out-of-range states, $0.69$ on scene-swapped states), and a diffusion decoder (PixNerd) reconstructs the camera frame from the latent, confirming the spatial compression preserves world-state. This report validates the single-timestep case; native-rate temporal fusion is the next step.
Chinese Translation
机器人必须理解自身的状态，但相机只能看到部分信息。力和接触在单帧中几乎没有痕迹，而在我们测试的每个机器人上，原始视觉特征在 $R^2$ 上的读取力低于或等于 $0.10$。我们提出了 Kepler-Encoder-v0.1，这是一种以机器人为中心的多模态编码器，将机器人状态视为一种模态，并通过一个学习查询的交叉注意力层将视觉、身体感知和力/扭矩融合为一个共享的潜在空间，采用基于遮罩的跨模态预测进行自监督训练，目标为 LeJEPA/SIGReg。在评估时，仅输入视觉，这提出了一个尖锐的问题。将状态融入训练是否使得仅基于视觉的潜在空间包含像素所不具备的内容？在 RH20T 语料库中，答案是肯定的，尤其是在相机最弱的地方。在保留的场景中，仅基于视觉的潜在空间显著恢复了末端执行器状态，尤其是力，远超原始的冻结 ViT 特征和每个传感器机器人上的计算匹配的仅视觉控制，尽管在单个时间步的绝对力恢复相对适中；在相机大部分可见的电机状态上，它的表现与最强的视觉基线在统计上相当，并且是唯一一个潜在几何跟踪状态的特征。一个与具体体现无关的编码器覆盖了四个机器人，而数据匹配的控制显示这种广度反映的是体现的多样性而非数据量。冻结的潜在空间直接有用。其自身的跨模态预测误差是一个无训练的无效状态监测器（在超范围状态下的 AUROC 为 $0.90$，在场景交换状态下的 AUROC 为 $0.69$），而扩散解码器（PixNerd）从潜在空间重建相机帧，确认空间压缩保留了世界状态。本报告验证了单时间步的情况；原生速率的时间融合是下一步。
cs.RO / 30 / 2607.13524
COLMAR: Cooperative View Policy Learning for Multi-Agent Active 3D Reconstruction
COLMAR：用于多智能体主动3D重建的协作视角策略学习
Abstract
Active 3D reconstruction requires selecting informative viewpoints under limited sensing budgets. In multi-agent settings, coordination inefficiencies such as redundant observations and spatial clustering can significantly reduce reconstruction quality. We present COLMAR, a cooperative view policy learning framework for multi-agent active 3D reconstruction. COLMAR formulates viewpoint allocation as a shared policy optimization over map-centric observations and introduces a reconstruction-aware objective that promotes overlap-aware coverage, team-level discovery, and collision-safe exploration. Dense feedback derived from incremental reconstruction updates aligns exploration behavior with downstream geometric quality. The policy is trained using parameter-sharing Proximal Policy Optimization (PPO) with independent per-agent action selection at deployment, conditioned on a fused team map and without inter-agent message passing for decision making. Selected viewpoints are then reconstructed with 3D Gaussian Splatting (3DGS) for high-fidelity photometric evaluation. Experiments on GLEAM and Replica demonstrate consistent improvements over heuristic and non-cooperative baselines, achieving up to 54% higher reconstruction accuracy and 49% greater coverage under matched sensing budgets.
Chinese Translation
主动3D重建需要在有限的感知预算下选择信息丰富的视角。在多智能体环境中，协调效率低下（如冗余观测和空间聚类）可能显著降低重建质量。我们提出了COLMAR，一个用于多智能体主动3D重建的协作视角策略学习框架。COLMAR将视角分配形式化为基于地图中心观测的共享策略优化，并引入了一种重建感知目标，促进重叠感知覆盖、团队级发现和安全探索。通过增量重建更新获得的密集反馈使探索行为与下游几何质量对齐。该策略使用参数共享的近端策略优化（Proximal Policy Optimization, PPO）进行训练，在部署时采用独立的每个智能体动作选择，基于融合的团队地图，并且在决策过程中不进行智能体间的信息传递。选定的视角随后使用3D高斯溅射（3D Gaussian Splatting, 3DGS）进行重建，以实现高保真度的光度评估。在GLEAM和Replica上的实验表明，与启发式和非协作基线相比，COLMAR在重建准确性上提高了多达54%，在匹配感知预算下覆盖率提高了49%。
cs.RO / 31 / 2607.13553
Flow-aware Optimal Navigation in Unsteady Flows through Reinforcement Learning
基于流动感知的强化学习在非稳态流动中的最优导航
Abstract
Autonomous robotic navigation in nonstationary time-varying fluid flows remains a fundamental challenge due to partial observability and the unpredictability of realistic environments. While classical optimal control frameworks employed in robotics require unrealistic a-priori global flow knowledge, biological systems are able to navigate successfully by exploiting localized sensory cues. In this work we present a reinforcement learning approach using the TD3 algorithm to train autonomous agents to reach arbitrary targets within a parametric, chaotic double-gyre flow. To investigate optimal sensory mechanisms, we evaluate five bio-inspired observation strategies based on relative position, local velocity or local vorticity measures, and short-term memory variants. Additionally, we analyze the impact of providing agents with explicit global flow parameters. Numerical results demonstrate that an agent that is able to sense and remember a set number of flow velocity measures achieves the highest performance. The experiments reveal a trade-off in sensor utility: velocity-aware agents optimize energy efficiency, whereas vorticity sensors provide superior structural mapping and achieve better target proximity. Incorporating explicit global flow parameters is shown to decrease navigation performance. This behavior suggests that reinforcement learning-based autonomous systems develop more robust and general policies when restricted to implicit flow representations. The presented results offer insights for improving the transition of bio-inspired robotic navigation from simulation to real-world environments.
Chinese Translation
在非静态时变流体流动中进行自主机器人导航仍然是一个基本挑战，这主要是由于部分可观测性和现实环境的不可预测性。虽然机器人学中采用的经典最优控制框架需要不切实际的先验全局流动知识，但生物系统能够通过利用局部感知线索成功导航。在本研究中，我们提出了一种使用TD3算法的强化学习方法，以训练自主代理在参数化的混沌双涡流中到达任意目标。为了研究最优感知机制，我们评估了五种基于相对位置、局部速度或局部涡度测量以及短期记忆变体的生物启发观察策略。此外，我们还分析了向代理提供显式全局流动参数的影响。数值结果表明，能够感知和记住一定数量流动速度测量的代理实现了最高的性能。实验揭示了传感器效用之间的权衡：速度感知代理优化了能量效率，而涡度传感器提供了更优的结构映射并实现了更好的目标接近度。引入显式全局流动参数被证明会降低导航性能。这种行为表明，基于强化学习的自主系统在限制于隐式流动表示时能够发展出更强大和更通用的策略。所呈现的结果为改善生物启发的机器人导航从模拟到现实环境的过渡提供了见解。
cs.RO / 32 / 2607.13573
IMMNet: Hybrid Fusion of Model-based and Data-driven Approaches for Maneuvering Target Tracking
IMMNet：基于模型与数据驱动方法的混合融合用于机动目标跟踪
Abstract
Maneuvering target tracking in three-dimensional space remains a challenging problem due to complex motion dynamics and model mismatch. To address this, this paper proposes a hybrid model/data-driven algorithm named IMMNet, which integrates the interpretable structure of the interacting multiple model (IMM) algorithm with learnable neural components. Unlike end-to-end black-box methods, the proposed IMMNet algorithm not only can preserve the Bayesian inference mechanism that is essential for real-time radar applications, but also can adaptively learn motion patterns and noise characteristics from data. Extensive experiments demonstrate that the proposed IMMNet algorithm consistently outperforms the existing algorithms across various scenarios, validating it as a robust, interpretable, and practical solution for maneuvering target tracking.
Chinese Translation
在三维空间中，机动目标跟踪仍然是一个具有挑战性的问题，主要由于复杂的运动动态和模型不匹配。为了解决这一问题，本文提出了一种名为IMMNet的混合模型/数据驱动算法，该算法将交互多模型（IMM）算法的可解释结构与可学习的神经组件相结合。与端到端的黑箱方法不同，所提出的IMMNet算法不仅能够保留对于实时雷达应用至关重要的贝叶斯推理机制，还能够自适应地从数据中学习运动模式和噪声特征。大量实验表明，所提出的IMMNet算法在各种场景中始终优于现有算法，验证了其作为机动目标跟踪的稳健、可解释和实用的解决方案。
cs.RO / 33 / 2607.13579
Agile perceptive multi-skill locomotion for quadrupedal robots in the wild
野外四足机器人灵活感知的多技能运动
Abstract
Enabling quadrupedal robots to traverse complex terrains-from rugged outdoor environments to urban landscapes-requires seamless integration of multiple motor skills, smooth transitions between gaits, and high-speed perceptive locomotion using only onboard sensors. We present APT-RL (Action Pretrained Transformer-based Reinforcement Learning), a unified framework that enables multi-skill locomotion to achieve high-speed traversal in complex environments through autonomous skill transitions utilizing only onboard perception and computation. Our approach generates large-scale, feature-rich 2D motion datasets through trajectory optimization with simplified dynamics. These datasets enable training of diverse, reusable locomotion skills that transfer effectively to a real quadruped robot operating on complex uneven terrains. The resulting high-quality skills serve as strong priors for efficient learning of complex downstream tasks and extend naturally to 3D environments, enabling smooth, high-speed multi-skill locomotion in deployed policy. Real-world experiments demonstrate the framework's capabilities: the robot performs agile maneuvers through complex indoor obstacles and outdoor wild environments, including dynamic drop-down maneuvers that reach instantaneous peak speeds of up to 6 meters per second. A single onboard policy enables robust traversal of diverse obstacles, including stairs, hurdles, stepping stones, gaps, and fallen branches, demonstrating the versatility and effectiveness of our approach.
Chinese Translation
使四足机器人能够在复杂地形中穿越——从崎岖的户外环境到城市景观——需要无缝整合多种运动技能、在步态之间平滑过渡，以及仅使用车载传感器进行高速感知运动。我们提出了APT-RL（基于动作预训练变换器的强化学习），这是一个统一框架，能够实现多技能运动，以通过仅依靠车载感知和计算实现复杂环境中的高速穿越。我们的方法通过简化动力学的轨迹优化生成大规模、特征丰富的二维运动数据集。这些数据集使得训练多样化、可重用的运动技能成为可能，这些技能能够有效转移到在复杂不平坦地形上操作的真实四足机器人上。生成的高质量技能为复杂下游任务的高效学习提供了强有力的先验，并自然扩展到三维环境中，使得在部署策略中实现平滑、高速的多技能运动成为可能。现实世界的实验展示了该框架的能力：机器人能够在复杂的室内障碍物和户外野外环境中进行灵活的机动，包括动态下落机动，瞬时峰值速度可达每秒6米。单一的车载策略使得机器人能够稳健地穿越多种障碍，包括楼梯、障碍物、踏脚石、缝隙和倒下的树枝，展示了我们方法的多样性和有效性。
cs.RO / 34 / 2607.13595
Active Trust Management for Successful Human-Robot Teaming: Moving from a Trust Repair to a Trust Satisficing Perspective
成功的人机团队信任管理：从信任修复到信任满意的视角
Abstract
Integrating mobile robots into human teams promises significant capability improvements for tasks such as searching hazardous environments. Unlike existing teleoperated robots, future robot systems will increasingly be endowed with some level of artificial intelligence (AI), giving them a degree of autonomy in how they pursue mission goals. This autonomy could make a human-agent (robot) team more effective but also put inter-agent trust under strain if robots make a mistake, or (appear to) pursue task priorities that conflict with the team's best interest. During a mission, agents' trust states are anticipated to vary according to the situation as understood by each teammate (trustor). If component-level (agent) or system-level trust falls below sufficient levels for cooperative tasks to be completed, it could critically affect mission success . We argue that active trust management will be an important precondition for the success of human-robot teams (HRTs, a subcategory of human-agent teams with embodied agents), especially in dynamic, high-risk environments. We present a trust satisficing perspective which acknowledges and attempts to account for the fluctuating, multi-faceted, and context-dependent nature of trust and trust requirements even under normal operating conditions. Our outline of a trust management framework for human-robot teaming includes online measurement of proxy metrics for trust, closed-loop adaptation of robot behavior, and variable autonomy to give space for human responsibility in situations requiring value judgements. We refer to a recent experimental exploration of 'swift trust' and a novel behavioral trust metric for HRT, and we highlight issues for further investigation.
Chinese Translation
将移动机器人整合到人类团队中，承诺在搜索危险环境等任务中带来显著的能力提升。与现有的遥控机器人不同，未来的机器人系统将越来越多地具备一定程度的人工智能（AI），使它们在追求任务目标时具有一定的自主性。这种自主性可能使人机（机器人）团队更有效，但如果机器人犯错，或（看似）追求与团队最佳利益相悖的任务优先级，则可能会对代理间的信任造成压力。在任务执行过程中，代理的信任状态预计会根据每个队友（信任者）理解的情况而变化。如果组件级（代理）或系统级信任降至完成合作任务所需的最低水平，可能会严重影响任务的成功。我们认为，积极的信任管理将是人机团队（HRTs，具体现身代理的人机团队子类别）成功的重要前提，特别是在动态的高风险环境中。我们提出了一种信任满意的视角，承认并试图考虑信任及其要求在正常操作条件下的波动性、多面性和情境依赖性。我们为人机团队信任管理框架的概述包括对信任的代理指标的在线测量、机器人行为的闭环适应以及可变自主性，以便在需要价值判断的情况下留出人类责任的空间。我们提及了对“快速信任”的近期实验探索和针对HRT的新型行为信任指标，并强调了进一步研究的问题。
cs.RO / 35 / 2607.13597
Semantic Anchoring for Robotic Action Representations
机器人动作表征的语义锚定
Abstract
Vision-Language-Action (VLA) models inherit rich semantic representations from pretrained Vision-Language Models, yet fine-tuning on limited robot demonstrations degrades this structure and undermines generalization. A fundamental question therefore arises: what constitutes a good action representation? Inspired by the mirror neuron theory's insight that observation and execution share an intention-level encoding, we examine whether a robot's action representations preserve the semantic structure captured by pretrained encoders. Systematic probing confirms that this structure erodes during finetuning, and that its quality synchronizes with both task success and out-of-distribution generalization. We further introduce a plug-and-play method that anchors action representations to a semantic manifold while decomposing representations into a shared semantic channel and a private channel, all discarded at inference, leaving the deployed model unchanged. Validated on different VLA backbones across simulation and real-world benchmarks, our method yields up to +18.7% on real-world in-distribution tasks and +21.5% on out-of-distribution generalization.
Chinese Translation
视觉-语言-动作（VLA）模型从预训练的视觉-语言模型中继承了丰富的语义表征，但在有限的机器人演示上进行微调会削弱这种结构并损害泛化能力。因此，一个基本问题随之而来：什么构成了良好的动作表征？受到镜像神经元理论的启发，该理论指出观察与执行共享意图层级编码，我们考察机器人的动作表征是否保留了由预训练编码器捕获的语义结构。系统性探测确认这种结构在微调过程中会逐渐消失，并且其质量与任务成功率和超出分布的泛化能力同步。我们进一步提出了一种即插即用的方法，将动作表征锚定到一个语义流形，同时将表征分解为一个共享语义通道和一个私有通道，所有这些在推理时被丢弃，从而保持已部署模型不变。在不同的VLA基础架构上进行的仿真和现实世界基准测试验证了我们的方法，在现实世界的同分布任务上提高了最高18.7%，在超出分布的泛化上提高了21.5%。
cs.RO / 36 / 2607.13605
An Empirical Study on Stage-Information Interfaces for VLA Fine-Tuning
关于 VLA 微调的阶段信息接口的实证研究
Abstract
One high-level instruction in long-horizon manipulation can cover several action stages. We use segmented action annotations as an intermediate representation between the full-task instruction and VLA action chunks. A progress module tracks the active stage, while the action policy receives stage information either as current-stage text or as a normalized ordinal stage index in robot state. We compare these interfaces with GR00T N1.6 on LIBERO-10 under direct fine-tuning and continuation fine-tuning from a full-task instruction baseline. Under direct fine-tuning, full-task instruction, current-stage text, and Ordinal Stage-State achieve mean success rates of 57.45%, 50.24%, and 54.36%, respectively, showing that explicit stage information does not automatically improve the policy. Under continuation, the corresponding means are 49.07%, 50.00%, and 53.75%, with Ordinal Stage-State exceeding both alternatives in all three paired runs. The observed benefit differs across interface representations and training arrangements.
Chinese Translation
在长时间操作中，一个高层次的指令可以涵盖多个动作阶段。我们使用分段动作注释作为完整任务指令与 VLA 动作块之间的中间表示。进度模块跟踪当前活动阶段，而动作策略则接收阶段信息，既可以是当前阶段的文本，也可以是机器人状态中的标准化序数阶段索引。我们在 LIBERO-10 上将这些接口与 GR00T N1.6 进行比较，分别在直接微调和从完整任务指令基线进行的继续微调下进行实验。在直接微调中，完整任务指令、当前阶段文本和序数阶段状态的平均成功率分别为 57.45%、50.24% 和 54.36%，显示出显式阶段信息并不自动改善策略。在继续微调中，相应的平均成功率为 49.07%、50.00% 和 53.75%，其中序数阶段状态在所有三次配对运行中均超过其他两种选择。观察到的效益在不同接口表示和训练安排之间存在差异。
cs.RO / 37 / 2607.13622
Design, Modeling and Experimental Validation of a Miniature Hybrid Underwater Glider With Large-Range Foldable Deflectable Wings
大型可折叠可偏转翼的微型混合水下滑翔器的设计、建模与实验验证
Abstract
Miniature hybrid underwater gliders have attracted increasing attention for long-endurance ocean observation and confined-space inspection. Large-range wing reconfiguration offers a promising yet largely unexplored approach for simultaneously enhancing maneuverability and shape adaptability in constrained underwater environments. However, such morphing introduces substantial challenges in mechanical integration, dynamic modeling, and hydrodynamic characterization. This paper presents FoDeGlider, a miniature hybrid underwater glider equipped with two independently actuated wings capable of large-range folding and deflection. To capture configuration-dependent variations in mass distribution, center-of-geometry location, and hydrodynamic loading, a multibody dynamics model is developed by treating wing configuration as a structural variable. A composite rigid body algorithm (CRBA)-based projection formulates the composite inertia, wrench transformations, and component-level hydrodynamics into a unified Fossen-form dynamic model applicable to arbitrary wing configurations. A sequential parameter-identification framework is further proposed to estimate fuselage and wing hydrodynamic coefficients, resulting in an open benchmark dataset for model identification and validation. Extensive experiments are conducted, the results of which demonstrate accurate dynamic modeling and parameter identification across diverse morphing configurations. Gate traversal experiments further validate FoDeGlider's ability to actively reconfigure its morphology during locomotion, enabling enhanced navigation in confined underwater environments.
Chinese Translation
微型混合水下滑翔器因其在长时间海洋观测和有限空间检查中的应用而受到越来越多的关注。大范围的翼型重构提供了一种有前景但尚未充分探索的方法，可以在受限的水下环境中同时增强机动性和形状适应性。然而，这种形态变化在机械集成、动态建模和流体动力特性方面引入了重大挑战。本文介绍了FoDeGlider，这是一种配备有两个独立驱动翼的微型混合水下滑翔器，能够进行大范围的折叠和偏转。为了捕捉与配置相关的质量分布、几何中心位置和流体动力加载的变化，本文开发了一种多体动力学模型，将翼的配置视为结构变量。基于复合刚体算法（CRBA）的投影将复合惯性、扭矩变换和组件级流体动力学整合为一个统一的Fossen形式动态模型，适用于任意翼配置。此外，提出了一种顺序参数识别框架，以估计机身和翼的流体动力系数，从而生成一个开放的基准数据集，用于模型识别和验证。进行了广泛的实验，结果表明在不同的形态变化配置下，动态建模和参数识别的准确性。门通行实验进一步验证了FoDeGlider在运动过程中主动重构其形态的能力，从而增强了在受限水下环境中的导航能力。
cs.RO / 38 / 2607.13624
From Language to Navigation Goals: A Vision-Language Approach for Semantic Navigation of Mobile Robots Using RGB-D Perception
从语言到导航目标：一种基于视觉-语言的移动机器人语义导航方法，利用RGB-D感知
Abstract
Natural language interaction provides an intuitive way for non-expert users to communicate with robotic platforms. However, transforming user requests into executable navigation actions remains a challenging task, requiring the integration of language understanding, environment perception, and autonomous navigation. This work presents a language-driven navigation framework that enables mobile robots to interpret user requests in natural language to move the robot to a destination and autonomously navigate towards it. The framework is composed of modular ROS 2 components that cooperate to transform natural language instructions into navigation actions. Given a natural language request referring to a target in the environment (e.g., "go to the mail box"), the system identifies the referenced object, estimates its position using RGB-D data, and generates a navigation goal, which is then executed through the ROS 2 Nav2 navigation stack. The ROS 2-based implementation facilitates portability across different robotic platforms, requiring only the configuration of the corresponding topics and services. The system is evaluated in both simulation and real-world scenarios using a TurtleBot3 Waffle and a Unitree Go2 robot with a RealSense camera. Experimental results show that the framework successfully interprets both direct commands and contextual requests, generates meaningful natural-language feedback, and navigates towards the desired target. These results demonstrate the feasibility of combining semantic perception and autonomous navigation to provide an intuitive human-robot interaction paradigm. Code will be released as open source upon acceptance.
Chinese Translation
自然语言交互为非专业用户与机器人平台之间的沟通提供了一种直观的方式。然而，将用户请求转化为可执行的导航动作仍然是一项具有挑战性的任务，这需要语言理解、环境感知和自主导航的整合。本研究提出了一种语言驱动的导航框架，使移动机器人能够解读用户的自然语言请求，以将机器人移动到目标位置并自主导航到达该位置。该框架由模块化的ROS 2组件组成，这些组件协同工作，将自然语言指令转化为导航动作。给定一个指向环境中目标的自然语言请求（例如，“去邮箱”），系统识别所提及的对象，利用RGB-D数据估计其位置，并生成导航目标，随后通过ROS 2 Nav2导航栈执行该目标。基于ROS 2的实现便于在不同的机器人平台之间移植，仅需配置相应的主题和服务。该系统在使用带有RealSense相机的TurtleBot3 Waffle和Unitree Go2机器人进行的仿真和现实场景中进行了评估。实验结果表明，该框架成功解读了直接命令和上下文请求，生成了有意义的自然语言反馈，并朝着期望的目标进行导航。这些结果展示了将语义感知与自主导航相结合以提供直观的人机交互范式的可行性。代码将在接受后作为开源发布。
cs.RO / 39 / 2607.13633
Design and Control of the "QuadBoat": A Quadruped Surface Vehicle for Drowning Rescue
“四足船”的设计与控制：一种用于溺水救援的四足水面车辆
Abstract
Prompt extraction of victims from water is crucial in water surface rescue missions. However, previous research on rescue robots has seldom addressed this issue. This paper presents QuadBoat, a bio-inspired unmanned surface vehicle (USV) designed to track and retrieve victims from water. QuadBoat features a quadrupedal robot configuration, enabling it with highly adaptable and agile maneuverability through its actively adjustable posture. Employing an inverse kinematics-based controller and cascaded model predictive control (MPC)-PID controller for overall movement, QuadBoat can accurately track and retrieve objects on the water surface. Maneuverability demonstrations validate QuadBoat's high agility, while a series of tracking experiments, including leg action tracking and trajectory tracking, confirm its high motion accuracy and system mobility. Finally, visual-based tracking and object pickup experiments further verify QuadBoat's target tracking capabilities and its effectiveness in executing rescues, both indoors and outdoors.
Chinese Translation
迅速从水中救出受害者在水面救援任务中至关重要。然而，以往关于救援机器人的研究很少涉及这一问题。本文介绍了QuadBoat，一种生物启发的无人水面车辆（USV），旨在追踪和救援水中的受害者。QuadBoat采用四足机器人配置，使其具备高度适应性和灵活的机动性，通过主动可调的姿态实现。QuadBoat采用基于逆运动学的控制器和级联模型预测控制（MPC）-PID控制器进行整体运动控制，能够准确追踪和救援水面上的物体。机动性演示验证了QuadBoat的高灵活性，而一系列追踪实验，包括腿部动作追踪和轨迹追踪，确认了其高运动精度和系统机动性。最后，基于视觉的追踪和物体拾取实验进一步验证了QuadBoat的目标追踪能力及其在室内和室外执行救援的有效性。
cs.RO / 40 / 2607.13696
Anatomy of Uncertainty: Expressive Descriptors of Robotic Manipulator Motion for Non-verbal Communication in Human-Robot Collaboration
不确定性的解剖：用于人机协作中非语言交流的机器人操控运动的表现性描述符
Abstract
Robots operating in human-robot collaboration must communicate not only their intended actions but also uncertainty arising from incomplete or ambiguous perception. This work introduces a mathematical framework for expressing perceptual uncertainty through robotic manipulator motion. Drawing on Laban Movement Analysis, robot behavior is organized in a Commitment-Vigilance state space that maps uncertainty-related states - confidence, curiosity, hesitance, fear, and inactivity - to distinct Laban Effort signatures. Five motion primitives - approach, pause, retreat, exploration, and oscillation - are then parameterized using eleven kinematic and geometric descriptors, including acceleration, pause and retreat characteristics, gaze angles, tilt, and shiver amplitude. A video-based human-subject study evaluated recognition of four expressive trajectories and the influence of individual descriptors on perceived intensity. Participants reliably identified the intended behavioral states, while several descriptors significantly modulated expressiveness. The results establish a perceptually grounded basis for encoding robot uncertainty in motion and support future autonomous trajectory generation using parametric movement representations for collaborative tasks in shared environments. Code, videos, questionnaire and appendices are available at "https://bit.ly/github-aou".
Chinese Translation
在进行人机协作的机器人必须不仅传达其预期的动作，还要表达由于感知不完整或模糊而产生的不确定性。本研究提出了一个数学框架，通过机器人操控运动来表达感知不确定性。基于拉班运动分析（Laban Movement Analysis），机器人行为被组织在一个承诺-警觉状态空间中，该空间将与不确定性相关的状态——信心、好奇心、犹豫、恐惧和不活动——映射到不同的拉班努力（Laban Effort）特征上。接着，五种运动原语——接近、暂停、撤退、探索和振荡——通过包括加速度、暂停和撤退特征、注视角度、倾斜度和颤动幅度在内的十一种运动学和几何描述符进行参数化。基于视频的人类受试者研究评估了四种表现性轨迹的识别及各个描述符对感知强度的影响。参与者可靠地识别了预期的行为状态，而多个描述符显著调节了表现性。研究结果为在运动中编码机器人不确定性建立了一个感知基础，并支持未来使用参数化运动表示进行自主轨迹生成，以便在共享环境中的协作任务中应用。代码、视频、问卷和附录可在 "https://bit.ly/github-aou" 获取。
cs.RO / 41 / 2607.13698
Towards a Modular Bin-picking Framework for Handling Object Pose Uncertainties
面向处理物体姿态不确定性的模块化拣箱框架
Abstract
In recent years, there has been growing interest in robust robotic systems for precise bin-picking applications. To achieve reliable performance, such systems must address errors arising from both the object pose estimation and the grasping process. Although various approaches have been proposed, they typically target specific challenges and do not offer general solutions. In this paper, we present a modular framework that jointly handles both error types. The framework incorporates object pose distribution estimation to account for pose uncertainty, which frequently arises in situations with ambiguous observations where a single correct pose cannot be determined. To further reduce uncertainty, we introduce a second-viewpoint module that computes complementary pose distributions, which are subsequently fused. This fusion decreases overall uncertainty and improves system efficiency. Additionally, two independent modules are included to compensate for grasping errors. The modular design allows the components to be combined for optimal performance or used individually, depending on the physical setup. The proposed method is evaluated in a real-world setup with three different objects, with no errors, and all modules are shown to improve efficiency. These results suggest that incorporating pose distributions with grasping pose errors is a promising direction for developing more flexible and reliable robotic production systems. To the best of our knowledge, this is the first framework that jointly addresses both grasping and object pose uncertainties using interchangeable modules. We believe there is ample opportunity to integrate additional modules, resulting in improved performance and flexibility. The current framework is limited to pose uncertainties in SO(2), but it could be extended to SE(3), enabling additional modules to improve the system.
Chinese Translation
近年来，针对精确拣箱应用的稳健机器人系统引起了越来越多的关注。为了实现可靠的性能，这类系统必须解决来自物体姿态估计和抓取过程的错误。尽管已有多种方法被提出，它们通常针对特定挑战，并未提供通用解决方案。本文提出了一种模块化框架，联合处理这两种错误类型。该框架结合了物体姿态分布估计，以考虑姿态不确定性，这种不确定性常常出现在观察模糊的情况下，无法确定单一正确的姿态。为了进一步减少不确定性，我们引入了一个第二视角模块，该模块计算互补的姿态分布，随后进行融合。这种融合降低了整体不确定性，提高了系统效率。此外，还包括两个独立模块以补偿抓取错误。模块化设计允许根据物理设置将组件组合以实现最佳性能，或单独使用。所提出的方法在一个真实环境中对三种不同物体进行了评估，未出现错误，所有模块均显示出提高效率的效果。这些结果表明，将姿态分布与抓取姿态错误结合起来是开发更灵活和可靠的机器人生产系统的一个有希望的方向。据我们所知，这是第一个使用可互换模块联合解决抓取和物体姿态不确定性的框架。我们相信还有很多机会可以集成额外模块，从而提高性能和灵活性。目前的框架仅限于SO(2)中的姿态不确定性，但可以扩展到SE(3)，以便增加额外模块来改善系统。
cs.RO / 42 / 2607.13704
nuTruck: Benchmarking Autonomous Driving Planning for Distributed Electric-drive Trucks
nuTruck：分布式电驱动卡车自主驾驶规划的基准测试
Abstract
The dominance of traditional rule-based methods in autonomous driving has gradually been replaced by learning-based approaches. While learning-based planners have achieved considerable success in passenger vehicles, their performance on heavy-duty trucks, particularly modern distributed electric-drive trucks (DETs), remains largely unexplored. To facilitate research and application of learning-based planners in DETs, this letter presents the first high-fidelity benchmark, called nuTruck, designed to support large-scale neural network training and closed-loop evaluation. Given the complex dynamics and high rollover susceptibility of DETs, we first incorporate a highly accurate nonlinear truck dynamical model into the simulation, which enables independent driving and steering of all wheels and captures dynamic load transfer caused by acceleration, deceleration, and cornering, thereby allowing quantitative assessment of rollover risk in closed-loop simulation. Second, we adapt several rule-based and learning-based planners as baselines for DETs and evaluate their performance in closed-loop simulation. Finally, using real-world driving scenarios from the nuPlan dataset, we conduct extensive closed-loop evaluations, analyzing not only conventional collision-free planning performance, but also the dynamical safety of the planned trajectories. The proposed nuTruck benchmark is expected to serve as a new standard for fair and realistic evaluation of autonomous driving planners on DETs.
Chinese Translation
传统基于规则的方法在自主驾驶领域的主导地位逐渐被基于学习的方法所取代。尽管基于学习的规划器在乘用车上取得了显著成功，但它们在重型卡车，特别是现代分布式电驱动卡车（DETs）上的表现仍然很大程度上未被探索。为了促进基于学习的规划器在DETs中的研究和应用，本文提出了第一个高保真基准，称为nuTruck，旨在支持大规模神经网络训练和闭环评估。考虑到DETs的复杂动态特性和高翻车敏感性，我们首先将一个高度准确的非线性卡车动力学模型纳入模拟中，该模型能够实现所有车轮的独立驱动和转向，并捕捉因加速、减速和转弯引起的动态载荷转移，从而允许在闭环模拟中对翻车风险进行定量评估。其次，我们将几种基于规则和基于学习的规划器调整为DETs的基准，并评估它们在闭环模拟中的表现。最后，利用来自nuPlan数据集的真实驾驶场景，我们进行了广泛的闭环评估，不仅分析了传统的无碰撞规划性能，还分析了规划轨迹的动态安全性。所提出的nuTruck基准预计将作为对DETs上的自主驾驶规划器进行公平和现实评估的新标准。
cs.RO / 43 / 2607.13768
The Nonsmooth Impact Direction (NSID) of Robotic Systems
机器人系统的非光滑冲击方向（NSID）
Abstract
Collisions of rigid-link robots and rigid environments are often modeled as instantaneous events. Under this idealization, the impact forces become impulsive and the system velocities nonsmooth. In this work, we systematically analyze pre- and post-impact velocities focusing on what we refer to as the nonsmooth impact direction (NSID). We show that it is a characteristic direction of a robotic impact and largely independent of contact properties. The results are directly applicable to large classes of backdrivable robotic systems with rigid links. We address particularities of systems with nonelastic and flexible joints, unconstrained as well as constrained systems. Further, we show that the approach direction w.r.t the NSID sets the direction of the impulsive force in frictional, inelastic impacts. The comprehensive theoretical analysis of this work supported by an experimental validation may serve as a foundation for future planning and control algorithms for various robotic impact applications. These can include humanoid locomotion on a slippery surface or repetitive hammering.
Chinese Translation
刚性连杆机器人与刚性环境的碰撞通常被建模为瞬时事件。在这种理想化下，冲击力变得具有冲击性，系统速度则呈现非光滑特性。在本研究中，我们系统地分析了冲击前后的速度，重点关注我们所称的非光滑冲击方向（NSID）。我们展示了它是机器人冲击的一个特征方向，并且在很大程度上独立于接触特性。该结果可直接应用于大类具有刚性连杆的可回驱动机器人系统。我们还讨论了具有非弹性和柔性关节的系统的特殊性，包括无约束和有约束系统。此外，我们表明，关于NSID的接近方向决定了摩擦性、非弹性冲击中的冲击力方向。本研究的全面理论分析及其实验验证可为未来各种机器人冲击应用的规划和控制算法奠定基础，这些应用包括在光滑表面上的类人运动或重复锤击。
cs.RO / 44 / 2607.13789
Dynamical Vehicle Orienteering Problem for Multi-Rotor Unmanned Aerial Vehicles
多旋翼无人机的动态车辆定向问题
Abstract
This paper introduces the Dynamical Vehicle Orienteering Problem (DVOP), a generalization of the Orienteering Problem (OP). The OP maximizes the reward collected from spatial targets under a limited travel budget; the DVOP extends it by accounting for both external and vehicle-actuated forces. We study the DVOP in the context of multi-rotor Unmanned Aerial Vehicle (UAV) flight planning, using a three-dimensional Point-Mass Model (PMM) constrained by maximum velocity and acceleration magnitudes and subject to gravitational acceleration, with the travel budget expressed as a maximum flight time. Because the DVOP couples reward maximization with time-optimal trajectory planning, it cannot be formulated as a simple graph problem and solved exactly without relaxing or under-actuating the vehicle dynamics. We therefore propose two solution approaches: a Branch-and-Bound (BnB) procedure that combines Non-Linear Programming (NLP) and Mixed-Integer Linear Programming (MILP) to provide high-quality solutions, and a Large Neighborhood Search (LNS) metaheuristic that supplies an initial reward bound and scales to instances intractable for the BnB. The BnB relies on a novel MILP formulation of travel costs based on minimum-time trajectory primitives through target triplets, yielding a tight reward upper bound, while the LNS uses limited thrust decomposition to compute fast, high-quality PMM trajectories. Experiments on benchmark instances show improvements of up to 37 % over state-of-the-art solutions for the Kinematic Orienteering Problem, and a real-world deployment on a multi-rotor UAV verifies the proposed PMM solution trajectories.
Chinese Translation
本文介绍了动态车辆定向问题（Dynamical Vehicle Orienteering Problem, DVOP），这是对定向问题（Orienteering Problem, OP）的推广。定向问题旨在在有限的旅行预算下最大化从空间目标收集的奖励；而动态车辆定向问题则通过考虑外部和车辆驱动的力量来扩展这一问题。我们在多旋翼无人机（Unmanned Aerial Vehicle, UAV）飞行规划的背景下研究DVOP，使用受最大速度和加速度限制的三维点质量模型（Point-Mass Model, PMM），并受到重力加速度的影响，旅行预算以最大飞行时间表示。由于DVOP将奖励最大化与时间最优轨迹规划结合在一起，因此无法简单地将其表述为图问题并精确求解，而不放宽或降低车辆动力学的约束。因此，我们提出了两种解决方案：一种是结合非线性规划（Non-Linear Programming, NLP）和混合整数线性规划（Mixed-Integer Linear Programming, MILP）的分支限界（Branch-and-Bound, BnB）程序，以提供高质量的解决方案；另一种是大邻域搜索（Large Neighborhood Search, LNS）元启发式算法，提供初始奖励界限，并适用于BnB无法处理的实例。BnB依赖于一种新颖的基于通过目标三元组的最小时间轨迹原语的旅行成本MILP公式，从而产生紧凑的奖励上界，而LNS则使用有限推力分解计算快速、高质量的PMM轨迹。在基准实例上的实验显示，相较于当前最先进的运动学定向问题解决方案，性能提高了多达37%，并且在多旋翼无人机上的实际部署验证了所提出的PMM解决方案轨迹。
cs.RO / 45 / 2607.13799
Vision-Based Obstacle Separation for Strawberry Harvesting in Clusters Using Hierarchical Reinforcement Learning
基于视觉的草莓聚集环境障碍物分离方法：层次强化学习框架
Abstract
Selective harvesting in clustered strawberry environments is challenging because ripe fruits are often occluded by surrounding unripe fruits, making direct grasping unreliable. To address this problem, this paper proposes a hierarchical reinforcement learning framework, termed VGPA, which integrates a vision-guided decision mechanism and a Progressive Adaptive Exploration Strategy (PAES) for vision-based obstacle separation and harvesting. The task was decomposed into two sequential stages: obstacle separation and target grasping. At the high level, the vision-guided mechanism improved option selection and accelerated policy convergence. At the low level, PAES improved exploration efficiency and training stability during continuous control learning. In simulation experiments, the learned policy achieved a success rate of 96.7%. In addition, sim-to-real transfer experiments on a self-developed parallel robot showed that the proposed method achieved success rates ranging from 71.7% to 88.3%, outperforming direct picking while requiring only 1.22~s more average harvesting time. These results verified the effectiveness, generalization ability, and practical potential of the proposed method for robotic harvesting in complex clustered environments.
Chinese Translation
在聚集的草莓环境中进行选择性采摘具有挑战性，因为成熟的果实常常被周围未成熟的果实遮挡，使得直接抓取变得不可靠。为了解决这个问题，本文提出了一种层次强化学习框架，称为VGPA，该框架集成了基于视觉的决策机制和渐进式自适应探索策略（Progressive Adaptive Exploration Strategy, PAES），用于视觉障碍物分离和采摘。该任务被分解为两个顺序阶段：障碍物分离和目标抓取。在高层次上，基于视觉的机制改善了选项选择并加速了策略收敛；在低层次上，PAES提高了连续控制学习中的探索效率和训练稳定性。在仿真实验中，学习到的策略达到了96.7%的成功率。此外，在自开发的并行机器人上的仿真到现实转移实验表明，所提方法的成功率在71.7%到88.3%之间，优于直接采摘，同时仅需多1.22秒的平均采摘时间。这些结果验证了所提方法在复杂聚集环境中进行机器人采摘的有效性、泛化能力和实际潜力。
cs.RO / 46 / 2607.13806
A Deployed Hybrid Vehicle-in-the-Loop Platform for Validating Cooperative Perception
一种部署的混合车辆环路平台用于验证协同感知
Abstract
European safety regulation now permits a large share of automated-driving homologation evidence to be produced virtually, provided a validated physical-virtual facility generates it. We present a deployed hybrid Vehicle-in-the-Loop (ViL) platform that couples a real instrumented vehicle with a CARLA-based digital twin (DT) through a V2X message pipeline, and we report its first integrated operation on a public-road-representative test track. A real vehicle streams ETSI-compliant CAM/CPM messages into the DT, where a GPU-accelerated Cooperative Perception (CP) module fuses them into a probabilistic occupancy grid during scenario runtime. We demonstrate the platform on a multi-vehicle double T-intersection scenario, characterise the CP workload across nominal, rain and night conditions and five localization-noise levels, and discuss the platform's current architectural limits and the engineering targets they define. The results show that CP substantially widens field-of-view (FoV) coverage and improves occupied-cell recall, and that beyond a moderate localization-noise threshold, positioning uncertainty, and not weather, becomes the dominant error source. We outline the platform's trajectory toward a Mediterranean operational design domain (ODD) testing service.
Chinese Translation
欧洲安全法规现在允许通过经过验证的物理-虚拟设施在虚拟环境中生成大量自动驾驶认证证据。我们提出了一种部署的混合车辆环路（Vehicle-in-the-Loop, ViL）平台，该平台通过V2X消息管道将一辆真实的仪器化车辆与基于CARLA的数字双胞胎（Digital Twin, DT）相结合，并报告其在公共道路代表性测试轨道上的首次集成操作。一辆真实车辆将符合ETSI标准的CAM/CPM消息流入DT，在场景运行期间，GPU加速的协同感知（Cooperative Perception, CP）模块将这些消息融合为概率占用网格。我们在一个多车辆双T型交叉口场景中演示了该平台，表征了在正常、雨天和夜间条件下以及五种定位噪声水平下的CP工作负载，并讨论了平台当前的架构限制及其定义的工程目标。结果表明，CP显著扩大了视场（Field-of-View, FoV）覆盖范围，并提高了占用单元的召回率，并且在超过适度的定位噪声阈值后，定位不确定性而非天气成为主要误差来源。我们概述了该平台朝向地中海操作设计域（Operational Design Domain, ODD）测试服务的轨迹。
cs.RO / 47 / 2607.13818
Learning Robust Execution in Robotic Manipulation with Agentic Reinforcement Learning
通过自主强化学习学习机器人操作中的稳健执行
Abstract
Robotic manipulation poses fundamental challenges due to uncertainty, long-horizon execution, and compounding errors, which can easily destabilize execution and lead to task failure. Although recent vision-language-action (VLA) models exhibit strong generalization, they typically lack explicit mechanisms to assess execution stability and to recover when execution deviates from its nominal behavior. In this paper, we propose: (1) two complementary metrics to assess execution quality at runtime, and (2) an agentic reinforcement learning framework that learns to restore effective execution through high-level decision-making rather than directly learning low-level actions. In this framework, an agentic policy reasons over recent execution history and selects among a small set of execution modes to regulate the execution process. Under execution degradation, it triggers appropriate recovery mechanisms to restore the robot to previously visited nominal states, enabling the task to continue. We evaluate the proposed method on the LIBERO benchmark, achieving up to a 13.7% improvement in success rate under standard settings and up to a 39.2% improvement under disturbance settings, demonstrating substantially enhanced execution robustness.
Chinese Translation
机器人操作面临着由于不确定性、长时间执行和累积错误而带来的基本挑战，这些因素很容易导致执行不稳定并导致任务失败。尽管最近的视觉-语言-动作（VLA）模型表现出强大的泛化能力，但它们通常缺乏评估执行稳定性和在执行偏离其正常行为时进行恢复的明确机制。本文提出：（1）两种互补的指标用于实时评估执行质量，以及（2）一个自主强化学习框架，该框架通过高层决策学习恢复有效执行，而不是直接学习低层动作。在该框架中，自主策略基于最近的执行历史进行推理，并在一小组执行模式中进行选择，以调节执行过程。在执行退化的情况下，它触发适当的恢复机制，将机器人恢复到先前访问的正常状态，从而使任务能够继续进行。我们在LIBERO基准上评估了所提出的方法，在标准设置下成功率提高了高达13.7%，在干扰设置下提高了高达39.2%，显示出显著增强的执行稳健性。
cs.RO / 48 / 2607.13830
Recursive ArUco Markers: A Scalable Fiducial Marker Design for Unmanned Aerial Vehicle Landing Pads
递归 ArUco 标记：一种可扩展的无人机着陆垫标记设计
Abstract
Unmanned Aerial Vehicles (UAVs) increasingly rely on visual fiducial markers for autonomous navigation and precision landing. However, standard markers suffer from limited operational ranges, becoming undetectable when the camera is either too far or too close. While recursive and fractal markers have been proposed to address this issue, existing approaches either require the marker's center to remain visible, making them vulnerable to occlusion, or are limited in their recursion depth and placement. We propose a novel Recursive ArUco marker design. Our method allows any standard fiducial marker to be transformed into a recursive marker with an arbitrary depth. By employing a modified bit-sampling strategy during detection, we embed complete markers within both the black and white bits of the parent marker. This approach guarantees unlimited recursion depth and robust detection even with partial occlusion, as it does not rely on the marker's center being visible. Furthermore, by maintaining a single, unique identifier across all recursive scales, our proposal provides an extensive dictionary of multiple unique landing pads. This capability allows fleets of UAVs to operate simultaneously, with each drone landing at its designated location -- a feature not supported by existing Fractal and Harco markers due to their structural and dictionary constraints.
Chinese Translation
无人机（UAV）越来越依赖视觉基准标记进行自主导航和精确着陆。然而，标准标记的操作范围有限，当相机距离过远或过近时，标记会变得不可检测。虽然已经提出了递归和分形标记来解决这一问题，但现有方法要么要求标记的中心保持可见，使其容易受到遮挡，要么在递归深度和放置上受到限制。我们提出了一种新颖的递归 ArUco 标记设计。我们的方法允许将任何标准基准标记转变为具有任意深度的递归标记。通过在检测过程中采用修改的位采样策略，我们将完整的标记嵌入到父标记的黑白位中。这种方法保证了无限的递归深度和即使在部分遮挡情况下也能进行稳健检测，因为它不依赖于标记中心的可见性。此外，通过在所有递归尺度上保持单一的唯一标识符，我们的提案提供了一个广泛的多个唯一着陆垫字典。这一能力使得无人机队能够同时操作，每架无人机在其指定位置着陆——这一特性是现有的分形和 Harco 标记由于其结构和字典限制所不支持的。
cs.RO / 49 / 2607.13853
Merging Reaction to Cognition: A Hybrid Cognitive Strategy for Odour Source Localisation in Natural Environments
反应与认知的融合：一种用于自然环境中气味源定位的混合认知策略
Abstract
Chemical pollutants released into the environment are transported by turbulent flows, generating complex, intermittent plume structures that threaten ecosystems and human health. Rapid localisation of emission sources is critical, and field robots equipped with chemical sensors provide a viable means to perform this task. However, inferring source location from sensor readings remains difficult due to sparse detections and the absence of reliable concentration gradients. Existing approaches fall into two paradigms. Bio-inspired strategies rely on reactive behaviours triggered by detections, such as surge-casting, offering efficiency but requiring scenario-specific tuning. Cognitive strategies integrate observations into a probabilistic belief over source location. While more robust, they suffer from excessive exploration and strong dependence on belief accuracy. The Fast-Cognitive algorithm reduced this computational burden but preserved the fundamental limitations. Previous Markov chain analysis revealed that source-directed motions occur roughly twice as often following odour detections, indicating that reactive behaviours naturally emerge within cognitive frameworks. This work proposes a hybrid strategy that explicitly incorporates bio-inspired reactivity into belief-dependent motion planning. It introduces a detection-triggered switching mechanism formalising transitions between crossflow exploration and source-directed motion, prioritising source proximity over information gain. Behavioural parameters are derived directly from belief metrics, enabling adaptive reactivity without manual tuning. The approach is validated through simulations under three turbulence conditions and field experiments with an autonomous surface vehicle in the Mondego River, Portugal. Results show up to 50% reduction in travelled distance, 86% success rate, and 3.2m average localisation error.
Chinese Translation
释放到环境中的化学污染物通过湍流传播，产生复杂的间歇性羽流结构，威胁生态系统和人类健康。快速定位排放源至关重要，配备化学传感器的场地机器人提供了一种可行的执行此任务的方法。然而，由于检测稀疏和缺乏可靠的浓度梯度，从传感器读数推断源位置仍然困难。现有方法分为两种范式。生物启发策略依赖于由检测触发的反应行为，如突发投射，提供了效率但需要特定场景的调优。认知策略将观察结果整合为对源位置的概率信念。尽管更为稳健，但它们面临过度探索和对信念准确性强依赖的问题。快速认知算法减少了这一计算负担，但保留了基本限制。先前的马尔可夫链分析表明，气味检测后，朝向源的运动发生的频率大约是正常情况的两倍，表明反应行为在认知框架中自然出现。本研究提出了一种混合策略，明确将生物启发的反应性纳入信念依赖的运动规划中。它引入了一种检测触发的切换机制，形式化了横流探索与朝向源运动之间的转换，优先考虑源的接近性而非信息增益。行为参数直接从信念指标中导出，使得在无需手动调优的情况下实现自适应反应。该方法通过在三种湍流条件下的仿真和在葡萄牙蒙德戈河的自主水面车辆的实地实验进行了验证。结果显示，行驶距离最多减少50%，成功率达到86%，平均定位误差为3.2米。
cs.RO / 50 / 2607.13882
Learning Forward & Reverse Skills from a Single Unfinished Demonstration for Constrained Manipulation Tasks
从单一未完成演示中学习约束操作任务的正向与反向技能
Abstract
Learning from demonstration (LfD) enables robots to learn manipulation skills directly from expert demonstrations but remains challenging for contact-rich tasks involving geometric constraints and force interaction. Existing approaches typically require multiple complete demonstrations and do not support reverse skill execution. In this paper, we present a unified one-shot framework for constrained manipulation that learns both forward and reverse execution from a single, possibly unfinished demonstration. Our method decomposes demonstrations into non-contact and contact phases, with non-contact motion encoded with dynamic movement primitives (DMP), and contact motion represented as a sequence of screw motion primitives segmented by our proposed geometry-driven twist-direction segmentation algorithm. During execution, screw primitives are executed sequentially under admittance-guided pose correction and speed regulation, enabling task completion beyond the demonstrated trajectory length as well as reverse skill execution without additional learning data. Experiments on peg insertion, battery insertion, lock opening, and screw driving tasks demonstrate improved success rates and robustness over segmentation and one-shot trajectory learning baselines. Details are available on the project website: https://tuwien-asl.github.io/LfD-Screw/.
Chinese Translation
通过演示学习（LfD）使机器人能够直接从专家演示中学习操作技能，但对于涉及几何约束和力交互的接触丰富任务仍然具有挑战性。现有的方法通常需要多个完整的演示，并且不支持反向技能执行。本文提出了一种统一的一次性框架，用于约束操作，能够从单个可能未完成的演示中学习正向和反向执行。我们的方法将演示分解为非接触和接触阶段，其中非接触运动使用动态运动原语（DMP）编码，而接触运动则表示为通过我们提出的基于几何的扭转方向分割算法分段的螺旋运动原语序列。在执行过程中，螺旋原语在导向性姿态修正和速度调节下顺序执行，使得任务完成超出演示轨迹的长度，并且能够在没有额外学习数据的情况下执行反向技能。在 peg 插入、电池插入、开锁和螺丝拧紧等任务上的实验表明，与分割和一次性轨迹学习基线相比，成功率和鲁棒性有所提高。详细信息可在项目网站上找到：https://tuwien-asl.github.io/LfD-Screw/.
cs.RO / 51 / 2607.13926
S-squared-VLA: Decoupling Semantic and Spatial Streams in Vision-Language-Action Models for Autonomous Driving
S平方-VLA：在自动驾驶的视觉-语言-动作模型中解耦语义和空间流
Abstract
Vision-Language Models (VLMs) have demonstrated remarkable potential for high-level reasoning in autonomous driving, yet they fundamentally struggle to generate precise, low-level control actions. This limitation is rooted in a semantic-physical gap caused by the inherent mismatch between discrete language tokens and continuous trajectory planning. While Vision-Language-Action (VLA) architectures attempt to bridge this gap by unifying perception and control into a single policy, this entanglement creates a new bottleneck. Standard VLAs experience a severe spatial representation collapse, which irreversibly degrades the fine-grained spatial and geometric priors essential for safe, boundary-aware navigation. To address this limitation, we propose the S-squared-VLA, which explicitly decouples the semantic and spatial streams in Vision-Language-Action models. The semantic stream leverages hierarchical bridging to extract multi-scale VLM features for robust intent reasoning. In parallel, an independent spatial stream bypasses the autoregressive language bottleneck, directly preserving uncompressed spatial features from the visual encoder. By integrating auxiliary perception supervision, this stream explicitly equips the model with rich spatial and geometric priors. Finally, a dual-stream planning adapter fuses high-level semantic intent with precise spatial constraints via cascaded attention mechanisms. Evaluations on the NAVSIM closed-loop benchmark show that S-squared-VLA achieves a Predictive Driver Model Score (PDMS) of 87.1, establishing a new state-of-the-art for VLA models under a purely supervised fine-tuning (SFT) setting. By mitigating the spatial representation collapse of traditional VLMs, our framework significantly outperforms baselines, achieving the highest No Collision (NC) rate of 98.4 among all evaluated methods.
Chinese Translation
视觉-语言模型（VLMs）在自动驾驶中的高层次推理方面展现了显著的潜力，但它们在生成精确的低层次控制动作方面存在根本性的困难。这一限制源于语义与物理之间的差距，这种差距是由于离散语言符号与连续轨迹规划之间的固有不匹配造成的。尽管视觉-语言-动作（VLA）架构试图通过将感知与控制统一为单一策略来弥合这一差距，但这种纠缠又创造了新的瓶颈。标准的VLA经历了严重的空间表示崩溃，这不可逆转地降低了安全且边界感知导航所必需的细粒度空间和几何先验。为了解决这一限制，我们提出了S平方-VLA，它明确地在视觉-语言-动作模型中解耦语义和空间流。语义流利用层次桥接提取多尺度VLM特征，以实现稳健的意图推理。与此同时，一个独立的空间流绕过自回归语言瓶颈，直接保留来自视觉编码器的未压缩空间特征。通过整合辅助感知监督，该流明确地为模型提供丰富的空间和几何先验。最后，一个双流规划适配器通过级联注意机制将高层次语义意图与精确的空间约束融合在一起。在NAVSIM闭环基准测试中的评估显示，S平方-VLA在纯监督微调（SFT）设置下达到了87.1的预测驾驶员模型分数（PDMS），为VLA模型建立了新的最先进水平。通过缓解传统VLM的空间表示崩溃，我们的框架显著优于基线，在所有评估方法中实现了98.4的最高无碰撞（NC）率。
cs.RO / 52 / 2607.13938
Discriminative Barrier Functions for Safe Adversarial Imitation Learning from Observation
用于安全对抗模仿学习的区分性障碍函数
Abstract
Inverse Reinforcement Learning (IRL) algorithms are powerful tools for learning from and generalizing expert demonstrations, but they often rely on unconstrained exploration, rendering them unsafe for real-world deployment. Meanwhile, Control Barrier Functions (CBFs) can guarantee the safety of control systems, but the analytical design of CBFs can be time-consuming and esoteric. In this work, we address these limitations jointly by constraining reward function candidacy during IRL to the space of CBFs, yielding a formulation that exhibits safe online control with continuous experiential improvement. Crucially, this framework enables the data-driven recovery of barrier functions directly from unlabeled expert observations. We demonstrate that the recovered barrier function is robust to unsafe states entirely absent from the expert data. Furthermore, we benchmark our method against standard IRL baselines in a simulated navigation environment, demonstrating improved safety performance. Finally, we investigate the trade-offs of planning-based versus policy-based IRL methods across both simulation and a real world obstacle avoidance task.
Chinese Translation
逆强化学习（IRL）算法是从专家演示中学习和推广的强大工具，但它们通常依赖于不受限制的探索，这使得它们在实际应用中不安全。同时，控制障碍函数（CBFs）可以保证控制系统的安全性，但CBFs的分析设计可能耗时且难以理解。在本研究中，我们通过在IRL中将奖励函数候选限制在CBFs的空间中，联合解决了这些限制，从而得出了一个在安全在线控制和持续经验改进方面表现良好的公式。关键是，这一框架能够直接从未标记的专家观察中数据驱动地恢复障碍函数。我们证明了恢复的障碍函数对完全不在专家数据中的不安全状态具有鲁棒性。此外，我们在一个模拟导航环境中将我们的方法与标准IRL基准进行了比较，展示了安全性能的改善。最后，我们研究了基于规划的方法与基于策略的IRL方法在模拟和现实世界障碍规避任务中的权衡。
cs.RO / 53 / 2607.13960
GigaWorld-Policy-0.5: A Faster and Stronger WAM Empowered by AutoResearch
GigaWorld-Policy-0.5：一种由AutoResearch驱动的更快更强的以行动为中心的世界行动模型
GigaWorld Team, Ye, Angen, Ma, Angyuan, Wang, Boyuan, Ni, Chaojun, Ye, Fangzheng, Huang, Guan, Li, Guo, Zhao, Guosheng, Yan, Haodong, Li, Hengtao, Lu, Jiwen, Wang, Kai, Yu, Mingming, Hu, Qitang, Deng, Qiuping, Liu, Songling, Tian, Xiaoyu, Wang, Xiaofeng, Zhou, Xinyu, Xu, Xiuwei, Chen, Xinze, Wang, Yang, Zeng, Yejun, Chang, Yifan, Ye, Yun, Wu, Zhenyu, Wu, Zhanqian, Zhu, Zheng
Abstract
World Action Models (WAMs) improve robot policy learning by jointly modeling actions and future visual observations, using future scene evolution as dense supervision for physically grounded action generation. However, a common design in existing WAMs is to explicitly generate future videos at inference time, incurring substantial computational overhead and hindering real-time closed-loop deployment. GigaWorld-Policy addresses this issue with an action-centered formulation, where future visual dynamics are used during training while action-only decoding is used at inference time. Building upon this framework, we present GigaWorld-Policy-0.5, an enhanced action-centered WAM designed for more efficient robot control. During pretraining, GigaWorld-Policy-0.5 adopts a mixed Action-Conditioned World Modeling (AC-WM) and WAM training strategy. This strengthens the coupling between visual dynamics and robot actions and improves the transferability of action representations for downstream policy learning. For efficient inference, GigaWorld-Policy-0.5 introduces a Mixture-of-Transformers architecture that separates visual dynamics modeling and action generation into specialized experts, reducing active computation during action-only inference and achieving 85 ms inference latency on a local RTX 4090 setup. In addition, we employ an agent-based AutoResearch pipeline to systematically search training configurations, enabling more efficient identification of optimal experimental setups while reducing the time and manual intervention required for hyperparameter tuning. Experiments and ablations show that GigaWorld-Policy-0.5 preserves the training benefits of future visual dynamics while improving inference efficiency for robot control.
Chinese Translation
世界行动模型（WAMs）通过共同建模动作和未来视觉观察来改善机器人策略学习，利用未来场景演变作为物理基础的动作生成的密集监督。然而，现有WAMs的一个常见设计是在推理时显式生成未来视频，这会产生大量计算开销，并阻碍实时闭环部署。GigaWorld-Policy通过一种以行动为中心的公式解决了这一问题，在训练过程中使用未来视觉动态，而在推理时仅使用动作解码。在此框架基础上，我们提出了GigaWorld-Policy-0.5，这是一种增强的以行动为中心的WAM，旨在实现更高效的机器人控制。在预训练期间，GigaWorld-Policy-0.5采用混合的动作条件世界建模（AC-WM）和WAM训练策略。这加强了视觉动态与机器人动作之间的耦合，并提高了动作表示在下游策略学习中的可转移性。为了实现高效推理，GigaWorld-Policy-0.5引入了一种混合变换器（Mixture-of-Transformers）架构，将视觉动态建模和动作生成分离为专门的专家，从而减少了在仅进行动作推理时的主动计算，并在本地RTX 4090设置上实现了85毫秒的推理延迟。此外，我们采用基于代理的AutoResearch管道系统地搜索训练配置，从而更高效地识别最佳实验设置，同时减少超参数调优所需的时间和人工干预。实验和消融研究表明，GigaWorld-Policy-0.5在提高机器人控制推理效率的同时，保留了未来视觉动态的训练优势。
cs.RO / 54 / 2607.14009
AeroMap3D: Anchoring Monocular UAV 6-DoF Localization to Visual-Geometric-Semantic Map Priors
AeroMap3D：将单目无人机6自由度定位锚定到视觉-几何-语义地图先验
Abstract
We present AeroMap3D, a monocular 6-DoF UAV localization system that anchors onboard imagery to visual, geometric, and semantic map priors for GNSS-denied navigation. AeroMap3D addresses two fundamental challenges in map-referenced aerial localization: the cross-view discrepancy between UAV imagery and satellite maps, and the structural inconsistency between bare-earth digital elevation models (DEMs) and urban scenes. First, we introduce a lightweight adapter that enables a dense matcher pretrained on internet-scale generic data to perform reliable UAV-to-map registration without finetuning. By estimating the scale ratio and yaw offset between the UAV image and map tile, the adapter removes the dominant geometric misalignment induced by altitude, camera field of view, and heading before dense correspondence estimation. Second, AeroMap3D lifts 2D UAV-map correspondences onto DEM terrain while using OpenStreetMap annotations to reject semantically unreliable matches before RANSAC-PnP pose estimation, thereby reducing errors caused by unmodeled building heights and off-nadir structures. Delayed map-based pose measurements are further fused with relative-motion priors using a delayed-state EKF for continuous trajectory estimation. Without UAV-Terra3D retraining or tuning, AeroMap3D localizes all trajectories across eight Austin sites within 50 m and achieves 5.88 m mean 3D error over 55 km of flight.
Chinese Translation
我们提出了AeroMap3D，这是一种单目6自由度无人机定位系统，能够将机载图像锚定到视觉、几何和语义地图先验，以实现无GNSS导航。AeroMap3D解决了地图参考的空中定位中的两个基本挑战：无人机图像与卫星地图之间的视角差异，以及裸土数字高程模型（DEM）与城市场景之间的结构不一致。首先，我们引入了一种轻量级适配器，使得在互联网规模的通用数据上预训练的密集匹配器能够在不进行微调的情况下可靠地执行无人机与地图的配准。通过估计无人机图像与地图切片之间的尺度比和偏航偏移，适配器在进行密集对应估计之前，消除了由高度、相机视场和航向引起的主要几何失配。其次，AeroMap3D将2D无人机-地图对应关系提升到DEM地形，同时利用OpenStreetMap注释在RANSAC-PnP姿态估计之前拒绝语义上不可靠的匹配，从而减少因未建模建筑高度和偏离天顶结构而导致的误差。延迟的基于地图的姿态测量进一步与相对运动先验通过延迟状态扩展卡尔曼滤波（EKF）融合，以实现连续轨迹估计。在不对无人机-Terra3D进行再训练或调优的情况下，AeroMap3D在八个奥斯汀地点的所有轨迹定位精度在50米以内，并在55公里的飞行中实现了5.88米的平均3D误差。
cs.RO / 55 / 2607.14021
Industrial Dexterity Benchmark: A Hardware-Software Benchmarking Platform for Industrial Dexterous Manipulation
工业灵巧性基准：用于工业灵巧操作的硬件-软件基准测试平台
Abstract
Dexterous manipulation remains a critical bottleneck in industrial automation; tasks such as cable routing, connector insertion, and precision assembly still rely heavily on manual labor despite decades of robotics research. This work presents a progression from classical, modular robotics pipelines toward an end-to-end multimodal imitation-learning framework for industrial dexterous manipulation. As a part of this work, we introduce three key contributions: a set of Industrial Dexterity Benchmark (IDB) boards aimed to mimic datacenter cable management, automotive cable harnesses, and gearbox assembly tasks; a scalable imitation learning framework (DAG-ROS); and a multimodal diffusion-based policy framework (AG-iDP3) that creates models fusing RGB images, point clouds, joint positions, and wrist-frame wrench data. Focusing on the datacenter cable manipulation board, we evaluate the performance of a task involving cleaning a single cable over variations of an end-to-end AI policy using 48 trials per configuration. The best performing configuration, a multimodal expansion Diffusion Policy (DP), includes a multi-view RGB image source passed through an R3M encoder and reaches a 78% grasp and insert combined task success rate. This performance marks a significant improvement over the 36% observed from the single-camera RGB DP baseline. Each of the tested configurations requires only approximately 100 teleoperated demonstrations per task phase. These results indicate that the correct learned policy can outperform classical vision and control robotic methods in robustness, generalization, and deployment efficiency, justifying a shift toward scalable robotic automation for high up-time industrial environments.
Chinese Translation
灵巧操作仍然是工业自动化中的一个关键瓶颈；尽管经过数十年的机器人研究，诸如电缆布线、连接器插入和精密组装等任务仍然在很大程度上依赖于人工劳动。本研究展示了从经典的模块化机器人管道向工业灵巧操作的端到端多模态模仿学习框架的进展。作为本研究的一部分，我们提出了三个关键贡献：一套旨在模拟数据中心电缆管理、汽车电缆线束和齿轮箱组装任务的工业灵巧性基准（IDB）板；一个可扩展的模仿学习框架（DAG-ROS）；以及一个基于多模态扩散的策略框架（AG-iDP3），该框架创建了融合RGB图像、点云、关节位置和腕部框架扭矩数据的模型。我们专注于数据中心电缆操作板，评估了涉及清理单根电缆的任务在48次试验中对端到端AI策略变体的性能。表现最佳的配置是多模态扩展扩散策略（DP），它包括通过R3M编码器传递的多视角RGB图像源，并达到了78%的抓取和插入综合任务成功率。这一性能相较于单摄像头RGB DP基线观察到的36%有了显著改善。每个测试配置仅需大约100次遥控演示即可完成每个任务阶段。这些结果表明，正确学习的策略在鲁棒性、泛化能力和部署效率上可以超越经典的视觉和控制机器人方法，从而证明了向高正常运行时间工业环境的可扩展机器人自动化转变的合理性。
cs.RO / 56 / 2607.14047
PhysClaw-0: A Symbiotic Agentic System for Robot Autonomy via Language Corrections
PhysClaw-0：通过语言修正实现机器人自主性的共生智能系统
Abstract
Autonomous data collection governs the volume and quality of real-world trajectories for manipulation policy learning. Existing pipelines reduce human effort via self-resetting, VLM verification, or language-guided correction, yet episode-scoped fixes must be reissued whenever the same failure recurs, so oversight cost grows with session length rather than with the number of distinct problems. We present PhysClaw-0, a human-robot symbiotic agentic system in which corrections are retained and reused across rounds. The collection loop collects, verifies, and resets autonomously, pausing for a remote operator only when a phase exhausts an explicit retry budget. An LLM parser maps each natural-language utterance to a structured adjustment stored in Corrective Memory, so addressed failure modes typically need not be corrected again under the same conditions. On a real-robot desktop-clearing testbed, PhysClaw-0 matches teleoperation episode success while reducing human working time to 16%. Language corrections improve verifier-human agreement in all four evaluated settings and raise average single-attempt success from 12.5% to 47.5% (arm-selection: 20.0% to 50.0%). Policies fine-tuned on PhysClaw-0 data match teleoperation-trained policy success at a fraction of collection human cost.
Chinese Translation
自主数据收集决定了用于操作策略学习的真实轨迹的数量和质量。现有的流程通过自我重置、VLM 验证或语言引导修正来减少人力投入，但每当相同的失败再次发生时，必须重新发出基于情节的修正，因此监督成本随着会话长度的增加而增加，而不是随着不同问题数量的增加。我们提出了 PhysClaw-0，这是一个人机共生智能系统，其中修正被保留并在多个回合中重复使用。收集循环自主进行收集、验证和重置，仅在某个阶段耗尽明确的重试预算时暂停以等待远程操作员。一个 LLM 解析器将每个自然语言表达映射到存储在修正记忆中的结构化调整，因此在相同条件下，已处理的失败模式通常不需要再次修正。在一个真实的机器人桌面清理测试平台上，PhysClaw-0 在匹配远程操作成功的同时，将人类工作时间减少至 16%。语言修正提高了四个评估环境中验证者与人类之间的一致性，并将单次尝试的平均成功率从 12.5% 提高到 47.5%（手臂选择：20.0% 提高到 50.0%）。在 PhysClaw-0 数据上微调的策略在收集人力成本的极小部分下，匹配了经过远程操作训练的策略成功率。
cs.CV / 1 / 2607.13116
C-Norm: Cell-Distribution Normalization Enables Precision Recognition of Medical-Cell Image
C-Norm：细胞分布归一化促进医学细胞图像的精确识别
Abstract
ThinPrep Cytologic Test (TCT) enables early cervical cancer screening, but manual reading is time-consuming and yields inconsistent diagnostic results among cytopathologists. Existing AI detection models perform poorly under real clinical conditions, primarily restricted by two key constraints: unbalanced spatial distribution of cell populations in TCT slides, and limited high-quality annotated cytology data relying on professional pathologist labeling. To address these limitations, we propose a Cell-Distribution Normalization (C-Norm) method. By decoupling abnormal and normal cells from the original TCT images and re-synthesizing them, this method ensures a uniform distribution of cell populations, thereby mitigating generalization degradation caused by distribution bias. Building upon this, we integrate the YOLOv12 framework with a DINOv3 module. This hybrid architecture leverages the advanced detection capability of YOLO models and the superior feature representations of DINOv3 to capture subtle morphological nuances essential for precise recognition of TCT images. Extensive experiments demonstrate that our proposed method achieves state-of-the-art performance, significantly outperforming mainstream detection algorithms. The complete implementation is available at: https://github.com/ddw2AIGROUP2CQUPT/Cell-Norm
Chinese Translation
ThinPrep细胞学检测（TCT）能够实现早期宫颈癌筛查，但人工阅读耗时且在细胞病理学家之间产生不一致的诊断结果。现有的人工智能检测模型在真实临床条件下表现不佳，主要受到两个关键限制的影响：TCT切片中细胞群体的空间分布不均衡，以及依赖专业病理学家标注的高质量细胞学数据有限。为了解决这些问题，我们提出了一种细胞分布归一化（C-Norm）方法。通过将异常细胞和正常细胞从原始TCT图像中解耦并重新合成，该方法确保细胞群体的均匀分布，从而减轻由分布偏差引起的泛化退化。在此基础上，我们将YOLOv12框架与DINOv3模块相结合。这种混合架构利用YOLO模型的先进检测能力和DINOv3的优越特征表示，以捕捉对TCT图像精确识别至关重要的细微形态特征。大量实验表明，我们提出的方法达到了最先进的性能，显著优于主流检测算法。完整实现可在以下网址获取：https://github.com/ddw2AIGROUP2CQUPT/Cell-Norm
cs.CV / 2 / 2607.13125
Boogu-Image-0.1: Boosting Open-Source Unified Multimodal Understanding and Generation
Boogu-Image-0.1：提升开源统一多模态理解与生成
Chen, Guoxuan, Xiao, Chufeng, Yang, Haoran, Xie, Siyue, Huang, Binxiao, Zhang, Ming, Chau, Cheuk Him, Fu, Xinyu, Lian, Yingzhao, Li, Tom S. Y., Lin, Jintao, Dong, Bowen, Qian, Zian, Liu, Yuhao, Hu, Yuxuan, Shi, Weikang, Zou, Bin, Zheng, Bowen, Che, Haoxuan, Chen, Chang, He, Yuyang, Sun, Heyang, Huang, Tianyu, Choi, Chong Hou, Gong, Cheng, Shi, Han, Bai, Haoli, Liu, Xihui, Li, Hongsheng, Chen, Qifeng, Huang, Chao, Liu, Rui, Lei, Chenyang
Abstract
We introduce Boogu-Image-0.1, an open-source unified multimodal understanding and generation model family, comprising Base, Turbo, Edit, and Edit-Turbo variants. It delivers competitive performance in high-quality text-to-image generation, fast inference, instruction-based editing, and bilingual (Chinese-English) text rendering. Closed-source multimodal systems like Nano-Banana-Pro and GPT-Image-2 achieve strong performance through system-level integration rather than a single model, yet their internal practices remain largely undisclosed. In this work, we demonstrate that targeted improvements in model understanding, data quality, and training pipelines, coupled with agentic inference-time scaling, can substantially enhance generation and editing performance even under highly constrained compute budgets. Comprehensive evaluations show that Boogu-Image-0.1 consistently matches or surpasses other open-source models across standard benchmarks, and achieves results approaching leading closed-source systems. Notably, this is accomplished with only 208.62 million unique images. The base model's theoretical training cost is only approximately \$400K. We share practical discussions that we believe are valuable to the broader research community, and release weights, code, and recipes under Apache 2.0 to advance the open ecosystem for unified multimodal understanding and generation. Our code is available here: https://github.com/Boogu-Project/Boogu-Image.
Chinese Translation
我们介绍了 Boogu-Image-0.1，这是一个开源的统一多模态理解与生成模型家族，包括 Base、Turbo、Edit 和 Edit-Turbo 变体。该模型在高质量文本到图像生成、快速推理、基于指令的编辑以及双语（中英文）文本渲染方面表现出色。与 Nano-Banana-Pro 和 GPT-Image-2 等闭源多模态系统相比，这些系统通过系统级集成而非单一模型实现了强大的性能，但其内部实践仍然在很大程度上未公开。在本研究中，我们展示了针对模型理解、数据质量和训练流程的有针对性的改进，加上推理时的代理性扩展，可以显著提升生成和编辑性能，即使在计算预算高度受限的情况下。全面的评估表明，Boogu-Image-0.1 在标准基准测试中始终与其他开源模型相匹配或超越，并且取得了接近领先闭源系统的结果。值得注意的是，这一切仅使用了 2.0862 亿张独特图像。基础模型的理论训练成本仅约为 40 万美元。我们分享了我们认为对更广泛研究社区有价值的实用讨论，并在 Apache 2.0 许可下发布权重、代码和配方，以推动统一多模态理解与生成的开放生态系统。我们的代码可在此获取：https://github.com/Boogu-Project/Boogu-Image。
cs.CV / 3 / 2607.13178
A Masked Autoencoder Approach to Unsupervised Steel Surface Defect Recognition
一种基于掩码自编码器的无监督钢铁表面缺陷识别方法
Abstract
Automated visual inspection of steel surface defects is a recurring quality control task in which labeled defect data is scarce and costly to obtain, while unlabeled surface images are abundant, which motivates self supervised methods that learn useful representations without class labels. A Transformer based Masked Autoencoder is used here to learn representations of steel surface defects for unsupervised grouping. During pretraining, 75% of the input image patches are randomly masked, and a lightweight decoder reconstructs the masked regions from the visible 25%. The encoder is trained jointly with an auxiliary defect localization objective, used only as a training signal and not evaluated as a detector. The decoder reaches a structural similarity score of 0.92 and a mean squared error of 0.47. Features from the pretrained encoder are then clustered using UMAP for dimensionality reduction and Agglomerative clustering, reaching a Hungarian matched accuracy of 91.3% against the six known defect categories.
Chinese Translation
钢铁表面缺陷的自动视觉检测是一个反复出现的质量控制任务，其中标记的缺陷数据稀缺且获取成本高，而未标记的表面图像则丰富，这促使了自监督方法的应用，这些方法在没有类别标签的情况下学习有用的表示。本文采用基于变换器的掩码自编码器（Masked Autoencoder）来学习钢铁表面缺陷的表示，以实现无监督分组。在预训练过程中，75%的输入图像块被随机掩盖，轻量级解码器从可见的25%重建被掩盖的区域。编码器与辅助缺陷定位目标共同训练，该目标仅作为训练信号使用，而不作为检测器进行评估。解码器达到了0.92的结构相似性评分和0.47的均方误差。然后，使用UMAP进行降维并结合聚合聚类对预训练编码器提取的特征进行聚类，针对六个已知缺陷类别达到了91.3%的匈牙利匹配准确率。
cs.CV / 4 / 2607.13187
MGFace: Mask-Gated Face Matching via Conditional Similarity Routing
MGFace：通过条件相似性路由的口罩门控人脸匹配
Abstract
Face identification has achieved remarkable performance under normal conditions. Yet, its accuracy often degrades significantly when query faces are partially occluded, especially by facial masks. Existing re-ranking approaches improve robustness by exploiting patch-level similarities. Still, they often rely on costly, fine-grained matching mechanisms, which limit their efficiency in large-scale retrieval scenarios. In this paper, we propose MGFace, a mask-gated face identification pipeline that predicts the mask status of a query face and conditionally routes the similarity computation accordingly. Specifically, MGFace distinguishes between masked and unmasked queries, applies global embedding matching to unmasked queries, and activates mask-aware patch-level re-ranking only for masked queries. This design focuses on reliable upper-face regions while avoiding unnecessary fine-grained computation. Experiments on the extended LFW-Mask dataset show that MGFace achieves over 80% identification accuracy with the FaceNet backbone and over 90% with the ArcFace backbone. Compared with a previous EMD-based re-ranking method, MGFace achieves better identification performance while reducing query time by approximately 20x. These results demonstrate the effectiveness of MGFace in improving masked-face identification accuracy with low computational overhead. The source code is available at https://github.com/chequanghuy/MGFace.
Chinese Translation
在人脸识别领域，在正常条件下已取得显著的性能。然而，当查询人脸部分被遮挡，尤其是被口罩遮挡时，其准确性往往显著下降。现有的重新排序方法通过利用局部相似性来提高鲁棒性，但它们通常依赖于昂贵的细粒度匹配机制，这限制了它们在大规模检索场景中的效率。本文提出了MGFace，一种口罩门控的人脸识别管道，该管道预测查询人脸的口罩状态，并相应地有条件地路由相似性计算。具体而言，MGFace区分有口罩和无口罩的查询，对无口罩查询应用全局嵌入匹配，仅对有口罩查询激活口罩感知的局部重新排序。该设计专注于可靠的上半脸区域，同时避免不必要的细粒度计算。在扩展的LFW-Mask数据集上的实验表明，MGFace在FaceNet骨干网络下实现了超过80%的识别准确率，在ArcFace骨干网络下实现了超过90%的识别准确率。与之前基于EMD的重新排序方法相比，MGFace在提高识别性能的同时，将查询时间减少了约20倍。这些结果证明了MGFace在提高口罩人脸识别准确性方面的有效性，同时保持低计算开销。源代码可在https://github.com/chequanghuy/MGFace获取。
cs.CV / 5 / 2607.13192
Self-Supervised Visual Representation Learning: Pretrain-Finetuning or Joint Training?
自监督视觉表征学习：预训练-微调还是联合训练？
Abstract
Self-supervision is a powerful technique for learning visual representations from unlabeled data. Existing techniques primarily adopt a two-stage approach for self-supervised learning (SSL): a pretraining stage on unlabeled data followed by a finetuning stage on labeled data. While this pipeline has demonstrated extreme effectiveness, the interaction between self-supervised and supervised learning objectives remains insufficiently understood. In this work, we systematically investigate whether jointly optimizing the self-supervised and supervised objectives during training provides a better alternative. We compare two training paradigms: (1) the aforementioned pretraining followed by finetuning (PFT) and (2) joint training (JT), where self-supervised and supervised losses are optimized simultaneously in the same network. Across eight representative SSL methods and diverse computer vision tasks on natural, medical, crisis response, and remote sensing data, we evaluate performance under varying percentages of labeled data. Our results reveal that the relative effectiveness of PFT and JT depends strongly on the task at hand, the availability of labeled data, and the complexity of the domain. We find that JT consistently improves data and training efficiency while being robust in low-label settings, while PFT is more reliable in more specialized domains. We further analyze representation quality, robustness, and cross-domain generalization, providing new insights into how self-supervised and supervised objectives interact during optimization. We establish a comprehensive empirical benchmark for hybrid SSL-based semi-supervised learning and offer practical guidance for selecting appropriate training strategies across diverse vision applications.
Chinese Translation
自监督是一种强大的技术，用于从未标记数据中学习视觉表征。现有技术主要采用两阶段的方法进行自监督学习（SSL）：在未标记数据上进行预训练，随后在标记数据上进行微调。尽管这一流程已显示出极高的有效性，但自监督学习与监督学习目标之间的相互作用仍然未得到充分理解。在本研究中，我们系统地探讨了在训练过程中是否通过联合优化自监督和监督目标可以提供更好的替代方案。我们比较了两种训练范式：（1）上述的预训练后跟微调（PFT），以及（2）联合训练（JT），在该方法中，自监督和监督损失在同一网络中同时优化。在八种代表性的SSL方法和自然、医学、危机响应及遥感数据等多样的计算机视觉任务中，我们评估了在不同标记数据比例下的性能。我们的结果表明，PFT和JT的相对有效性在很大程度上依赖于具体任务、标记数据的可用性以及领域的复杂性。我们发现，JT在低标记设置下始终提高了数据和训练效率，同时表现出鲁棒性，而PFT在更专业的领域中更为可靠。我们进一步分析了表征质量、鲁棒性和跨领域泛化，为自监督和监督目标在优化过程中的相互作用提供了新的见解。我们建立了一个全面的经验基准，用于基于混合SSL的半监督学习，并为在多样化视觉应用中选择合适的训练策略提供了实用指导。
cs.CV / 6 / 2607.13234
Continuously Evolving Deepfake Detection: An Architecture and Public-Benchmark Evaluation of a Dynamic Detection System
持续演变的深伪检测：动态检测系统的架构与公共基准评估
Abstract
Deepfake detectors that achieve near-perfect scores on academic benchmarks collapse on real-world content: recent in-the-wild evaluations report AUC drops of 45-50% for state-of-the-art open-source models. We argue this gap is structural: static detectors are trained once against a moving generative frontier. We present BitMind Forensics (BMF), trained through Bittensor SN34, an open adversarial competition that continually refreshes the training distribution. We evaluate one dated export comprising image, general-video, and human-video checkpoints across nineteen public datasets: the canonical face-swap suites (FaceForensics++, Celeb-DF v1/v2/++, DFDC, DFD, UADFV, DF40) and recent in-the-wild and AI-generated-media benchmarks (Sumsub, Deepfake-Eval-2024, WildRF, Community Forensics, AIGCDetectBench, GenImage, AI-GenBench, AIGIBench, RAID, GenVidBench, GenVideo-100K). BMF reaches 0.936 AUC on Sumsub's original images and 0.872 pooled AUC over its full four-condition manipulation battery (1.4M images), staying robust under perturbation (0.855 JPEG, 0.799 downscaled), while GPEN enhancement improves detection (0.996). On Deepfake-Eval-2024, it matches the best commercial detector on images (0.915 vs 0.90) and exceeds it on video (0.822 vs 0.79), far above the best open-source detectors (0.56 and 0.63). It reaches 0.991 AUC on a 21-generator AI-image panel and 0.918 on GenVidBench, and exceeds the FF++-trained frontier on DFDC (0.947 vs 0.843) and Celeb-DF v2 (0.9985 vs 0.956), both contamination-audited, with statistical parity on Celeb-DF++. In a temporal study, successive dated exports improve on held-out media from generators absent from the static baseline's training (image 0.842 to 0.902; video 0.864 to 0.936). Our evaluation harness is public, and at publication the production API serves the exact evaluated snapshot for independent verification.
Chinese Translation
在学术基准上取得近乎完美分数的深伪检测器在现实世界内容上却表现不佳：最近的实地评估报告显示，最先进的开源模型的AUC下降了45-50%。我们认为这种差距是结构性的：静态检测器是在不断变化的生成前沿上进行一次性训练的。我们提出了BitMind Forensics (BMF)，通过Bittensor SN34进行训练，这是一个持续更新训练分布的开放对抗竞赛。我们评估了一个过时的导出，包括在十九个公共数据集上的图像、一般视频和人类视频检查点：经典的换脸套件（FaceForensics++、Celeb-DF v1/v2/++、DFDC、DFD、UADFV、DF40）以及最近的实地和AI生成媒体基准（Sumsub、Deepfake-Eval-2024、WildRF、Community Forensics、AIGCDetectBench、GenImage、AI-GenBench、AIGIBench、RAID、GenVidBench、GenVideo-100K）。BMF在Sumsub的原始图像上达到了0.936的AUC，在其完整的四种条件操作电池（140万张图像）上达到了0.872的汇总AUC，并在扰动下保持稳健（JPEG为0.855，降级为0.799），同时GPEN增强提高了检测能力（0.996）。在Deepfake-Eval-2024上，它在图像上与最佳商业检测器相匹配（0.915对0.90），在视频上超过了它（0.822对0.79），远高于最佳开源检测器（0.56和0.63）。在一个21个生成器的AI图像面板上，它达到了0.991的AUC，在GenVidBench上达到了0.918，并在DFDC上超过了FF++训练的前沿（0.947对0.843）和Celeb-DF v2（0.9985对0.956），两者均经过污染审计，并在Celeb-DF++上实现了统计平等。在一项时间研究中，连续的过时导出在来自静态基线训练中缺失的生成器的保留媒体上有所改善（图像从0.842提高到0.902；视频从0.864提高到0.936）。我们的评估工具是公开的，在发布时，生产API提供了精确评估快照以供独立验证。
cs.CV / 7 / 2607.13237
Active Learning for Efficient Annotation of Surgical Videos with Weak Supervision
基于弱监督的手术视频高效标注的主动学习
Abstract
Precise spatial-temporal annotation of laparoscopic videos is time-consuming and requires expert knowledge. We propose a human-in-the-loop knowledge acquisition framework that combines active learning with dual-loss optimization to significantly reduce the annotation effort needed for automatic localization and segmentation of objects in the surgical field. Our method employs a foundation model to generate temporally consistent class activation maps (CAMs) from video using two complementary training objectives: a weak supervision loss on video-level tool presence labels for weakly annotated data, and an image-level mask loss on human-corrected annotations obtained through active learning. Rather than requiring dense pixel-level annotation upfront, our pipeline iteratively proposes pseudo-masks that guide the expert annotator to refine the knowledge previously captured by the model. We demonstrate that our framework reduces the effort of surgical video annotation by 50% by the end of training in comparison to fully manual annotation. Through eliminating the need for large, fully annotated datasets from the start, this framework enables scalability to the development of surgical tool segmentation models. This iterative human-in-the-loop refinement supports efficient knowledge acquisition with minimal expert input, providing a practical and deployable strategy for expanding tool segmentation to larger, more diverse datasets and real-world clinical settings.
Chinese Translation
腹腔镜视频的精确时空标注耗时且需要专业知识。我们提出了一种人机协作的知识获取框架，该框架结合了主动学习与双损失优化，以显著减少自动定位和分割手术领域中物体所需的标注工作。我们的方法采用基础模型，从视频中生成时序一致的类别激活图（Class Activation Maps, CAMs），使用两个互补的训练目标：针对弱标注数据的基于视频级工具存在标签的弱监督损失，以及通过主动学习获得的人类修正标注的图像级掩码损失。我们的流程并不需要一开始就进行密集的像素级标注，而是迭代地提出伪掩码，引导专家标注者细化模型之前捕获的知识。我们证明，与完全手动标注相比，我们的框架在训练结束时将手术视频标注的工作量减少了50%。通过消除从一开始就需要大规模完全标注数据集的需求，该框架使得手术工具分割模型的开发具有可扩展性。这种迭代的人机协作精炼支持以最小的专家输入实现高效的知识获取，为将工具分割扩展到更大、更具多样性的数据集和现实临床环境提供了一种实用且可部署的策略。
cs.CV / 8 / 2607.13245
Just-In-Time Scene Graph Growth: Combating Perceptual Saturation in Long-Horizon Robotics
即时场景图增长：应对长时间跨度机器人中的感知饱和
Abstract
While 3D Scene Graphs (3DSGs) provide crucial structured representations for embodied agents, conventional Ahead-of-Time, build-everything-then-filter pipelines conflict with the real-time, low-latency demands of edge platforms, inducing a perceptual saturation effect via severe observation redundancy. To resolve this, we present JITOMA (Just-In-Time On-demand Memory Activation), a closed-loop framework that unifies task reasoning, perception, and memory into a just-in-time growth process. Instead of exhaustively mapping the entire environment, JITOMA leverages a top-down task heatmap at the frontend to filter continuous observations, routing minimal streams to maintain a global foundation of low-cost, dormant anchors. Upon a cognitive query, the backend Large Language Model (LLM) parses the robotic intent to dynamically awaken task-relevant anchors, triggering resource-intensive operations -- such as dense node captioning and functional inference -- exclusively within the activated local subgraph. To evaluate these dynamic capabilities and study perceptual saturation trade-offs, we introduce JITOMA-Bench, a comprehensive suite for long-horizon multi-tasking and complex multi-step reasoning. Extensive experiments demonstrate that JITOMA substantially reduces active graph size and captioning latency, while maintaining stable processing time under long-horizon task switching.
Chinese Translation
尽管3D场景图（3DSGs）为具身智能体提供了重要的结构化表示，但传统的提前构建、先生成所有内容再过滤的流程与边缘平台对实时性和低延迟的需求相冲突，导致严重的观察冗余，从而引发感知饱和效应。为了解决这一问题，我们提出了JITOMA（即时按需内存激活），这是一个将任务推理、感知和记忆统一为即时增长过程的闭环框架。JITOMA并不是全面映射整个环境，而是在前端利用自上而下的任务热图来过滤连续观察，将最小的数据流路由到保持低成本、休眠锚点的全球基础上。在认知查询时，后端的大型语言模型（LLM）解析机器人的意图，以动态唤醒与任务相关的锚点，触发资源密集型操作——如密集节点标注和功能推理——仅在激活的局部子图内进行。为了评估这些动态能力并研究感知饱和的权衡，我们引入了JITOMA-Bench，这是一个用于长时间跨度多任务和复杂多步骤推理的综合套件。大量实验表明，JITOMA显著减少了活动图的大小和标注延迟，同时在长时间跨度任务切换下保持稳定的处理时间。
cs.CV / 9 / 2607.13250
AffectFlow-DINO: Uncertainty-Aware Multi-Task Affect Estimation via Conditional Rectified Flow
AffectFlow-DINO：通过条件整流流实现的不确定性感知多任务情感估计
Abstract
We present \textbf{AffectFlow-DINO}, a multi-task learning system for the 11th ABAW challenge that extends a standard deterministic architecture with a conditional rectified-flow head to model the inherent ambiguity of in-the-wild facial behavior. Instead of predicting a single affect estimate, the model learns a conditional generative distribution, enabling uncertainty-aware one-to-many predictions through Monte Carlo sampling. The system jointly estimates continuous valence-arousal, classifies eight facial expressions, and detects twelve Action Units from static face images. Built on a frozen DINOv3 ViT-S/16 backbone, extensive ablation studies show that rectified-flow decoding consistently improves deterministic prediction, particularly for valence-arousal estimation (CCC-V $+0.058$). We further show that post-hoc threshold calibration effectively recovers performance on severely imbalanced rare classes (e.g., Fear: $3.8\% \rightarrow 33.1\%$) without retraining. Combined with backbone fine-tuning and flow retuning, the final model achieves $\mathbf{P_{MTL}=1.177}$, substantially outperforming the official challenge baseline of $P_{MTL}=0.45$.
Chinese Translation
我们提出了 extbf{AffectFlow-DINO}，这是一个多任务学习系统，旨在参加第11届ABA挑战赛，该系统通过条件整流流头扩展了标准的确定性架构，以建模自然环境中面部行为的内在模糊性。该模型不是预测单一的情感估计，而是学习一个条件生成分布，从而通过蒙特卡罗采样实现不确定性感知的一对多预测。该系统共同估计连续的愉悦-唤醒（valence-arousal），分类八种面部表情，并从静态面部图像中检测十二个动作单元（Action Units）。基于冻结的DINOv3 ViT-S/16主干网络，广泛的消融研究表明，整流流解码始终改善确定性预测，特别是在愉悦-唤醒估计方面（CCC-V $+0.058$）。我们进一步展示了后验阈值校准有效地恢复了在严重不平衡的稀有类别上的性能（例如，恐惧：$3.8\%
ightarrow 33.1\\%$），而无需重新训练。结合主干微调和流重调，最终模型达到了$ extbf{P_{MTL}=1.177}$，显著超越了官方挑战基线$P_{MTL}=0.45$。
cs.CV / 10 / 2607.13265
Differentiable Polarized Path Tracing
可微分偏振路径追踪
Abstract
Physically based differentiable rendering has proven to be a powerful tool for inverse rendering problems (e.g., 3D reconstruction, reflectance estimation, lighting estimation). However, most existing methods operate solely on radiometric intensity, discarding valuable polarization cues that constrain scene geometry and material properties. While forward simulation of polarized light is well-defined via Mueller-Stokes calculus, extending reverse-mode differentiation to this domain presents significant challenges. The rank-deficient nature of common polarimetric operators, such as linear polarizers and diffuse reflections, violates the invertibility assumptions of standard gradient estimators like path replay backpropagation and results in numerical instability. We address this by proposing a robust, polarization-aware differentiable path tracing method. Our approach estimates unbiased gradients through a combination of path replay and local caching. This formulation enables efficient and stable optimization of material and lighting parameters in complex scenes, broadening the applicability of physically based inverse rendering. Project page: https://vcai.mpi-inf.mpg.de/projects/DPPT/
Chinese Translation
基于物理的可微分渲染已被证明是解决逆渲染问题（例如，3D重建、反射率估计、光照估计）的强大工具。然而，大多数现有方法仅基于辐射强度，忽略了宝贵的偏振线索，这些线索可以约束场景几何和材料特性。虽然通过穆勒-斯托克斯（Mueller-Stokes）微积分对偏振光的正向模拟有明确的定义，但将反向模式微分扩展到这一领域面临重大挑战。常见的偏振测量操作符（如线性偏振器和漫反射）的秩缺陷特性违反了标准梯度估计器（如路径重放反向传播）的可逆性假设，导致数值不稳定。我们通过提出一种稳健的、考虑偏振的可微分路径追踪方法来解决这一问题。我们的方法通过路径重放和局部缓存的组合来估计无偏梯度。这种公式化方法使得在复杂场景中有效且稳定地优化材料和光照参数成为可能，拓宽了基于物理的逆渲染的适用性。项目页面：https://vcai.mpi-inf.mpg.de/projects/DPPT/
cs.CV / 11 / 2607.13298
FOLIO: Focused Semantic Memory for Streaming Video Understanding
FOLIO：用于流媒体视频理解的聚焦语义记忆
Abstract
In online streaming video understanding, a video stream continues to arrive and queries may be issued at any time. Because streaming frames grow without bound, the system must continuously compress and retain information from the observed video prefix while future frames and future queries remain unknown. The core challenge is deciding what information to retain and how to organize the maintained history: as this history grows with the stream, memory cost increases and many redundant visual details are retained, whereas later queries often depend on specific entities, actions, and their temporal changes. To address this challenge, we introduce FOLIO, a training-free focused semantic memory system that records important parts of the stream in higher detail while keeping surrounding context compact. As the stream arrives, FOLIO updates memory at the segment level, guided by a dynamic focus state, combining a short-term visual buffer with a long-term semantic memory organized around observed entities and linked to a visual-evidence cache. At query time, lightweight hybrid retrieval combines direct matching over the structured memory with semantic query expansion. FOLIO achieves state-of-the-art performance, reaching 82.0/69.1 Perception/Backward accuracy on OVO-Bench with Qwen3-VL-8B and 74.5 overall accuracy on StreamingBench, while substantially reducing the cost of maintaining streaming memory by reserving detailed records for focused entities and storing surrounding context compactly.
Chinese Translation
在在线流媒体视频理解中，视频流不断到达，查询可能随时发出。由于流媒体帧数量无限增长，系统必须持续压缩并保留观察到的视频前缀中的信息，同时未来的帧和查询仍然未知。核心挑战在于决定保留哪些信息以及如何组织维护的历史：随着历史随着流的增长，内存成本增加，许多冗余的视觉细节被保留，而后续查询往往依赖于特定的实体、动作及其时间变化。为了解决这一挑战，我们提出了FOLIO，一种无训练的聚焦语义记忆系统，它以更高的细节记录流的重要部分，同时保持周围上下文的紧凑。当流到达时，FOLIO在段级别更新内存，受动态聚焦状态的指导，将短期视觉缓冲区与围绕观察到的实体组织的长期语义记忆结合，并链接到视觉证据缓存。在查询时，轻量级混合检索结合了对结构化内存的直接匹配与语义查询扩展。FOLIO实现了最先进的性能，在OVO-Bench上以Qwen3-VL-8B达到了82.0/69.1的感知/向后准确率，并在StreamingBench上达到了74.5的整体准确率，同时通过为聚焦实体保留详细记录并紧凑存储周围上下文，显著降低了维护流媒体记忆的成本。
cs.CV / 12 / 2607.13300
Improving Medical Image Generative Models with Fr\'echet Distance Loss
通过Fréchet距离损失改进医学图像生成模型
Abstract
Diffusion generative models have demonstrated immense potential for synthetic medical image generation. However, these models often struggle to capture complex morphological characteristics of heterogeneous tumors with irregular boundaries, limiting their utility for downstream clinical tasks such as segmentation. This limitation stems from the standard denoising objective: minimizing a per-pixel error, which smooths high-variance irregular structures characteristic of tumors. To address this, we propose finetuning these generative models with Fr\'echet Distance loss (FD-loss). FD-loss aligns the first and second order feature statistics of real and generated images in a pretrained encoder space, encouraging the generator to capture complex structural variations characteristic of heterogeneous tumors. We integrate FD-loss across diverse architectural settings, using both natural- and medical-image encoders on multiple liver and brain cancer datasets spanning CT and MRI modalities. Downstream segmentation networks trained on our FD-regularized synthetic data consistently achieve superior performance, improving tumor DSC by $>$$5\%$ over unregularized synthetic augmentation alone. Qualitative analysis suggests these gains are associated with more faithful tumor synthesis and fewer segmentation hallucinations. Our results show FD-loss as an effective regularizer for medical image generative models to improve clinical workflows.
Chinese Translation
扩散生成模型在合成医学图像生成方面展现了巨大的潜力。然而，这些模型通常难以捕捉具有不规则边界的异质肿瘤的复杂形态特征，从而限制了它们在下游临床任务（如分割）中的应用。这一限制源于标准去噪目标：最小化每个像素的误差，这会平滑肿瘤特征的高方差不规则结构。为了解决这一问题，我们提出使用Fréchet距离损失（FD-loss）对这些生成模型进行微调。FD-loss在预训练编码器空间中对真实图像和生成图像的第一和第二阶特征统计进行对齐，鼓励生成器捕捉异质肿瘤特征的复杂结构变化。我们在多种架构设置中整合FD-loss，使用自然图像和医学图像编码器，针对多个肝脏和脑癌数据集（涵盖CT和MRI模态）进行实验。在我们的FD正则化合成数据上训练的下游分割网络始终实现了更优的性能，相比于仅使用未正则化的合成增强，肿瘤的DSC提高了超过5%。定性分析表明，这些提升与更真实的肿瘤合成和更少的分割幻觉相关。我们的结果表明，FD-loss是医学图像生成模型的有效正则化器，有助于改善临床工作流程。
cs.CV / 13 / 2607.13305
Accuracy Without Grounding: Diagnosing Visual Dependency Dissociation in Video LLM Benchmarks
无基础的准确性：诊断视频大语言模型基准中的视觉依赖解离
Abstract
Benchmark accuracy in video large language models (LLMs) is often treated as evidence of visual understanding. We audit this assumption across twenty models spanning 2-78B parameters and ten architecture families. We introduce the Visual Dependency Gap (VDG), the difference in per-question correctness between original-video and black-screen conditions. Paired McNemar tests on MVBench show that accuracy and visual dependency are separable: models differ on original video (p = 0.0003) but not on black screens (p = 0.53). Across models, task-type rankings are stable: Attribute Perception is strongly visual, whereas Temporal Reasoning approaches the language-only baseline. A diagnostic ladder from black screen to single frame, shuffled frames, and original video reveals that frame diversity supplies most of the visual benefit, while temporal order contributes near-zero accuracy across sixteen open-weight models. An ablation from 0.5 to 24 FPS rules out sparse sampling as the cause. H.264 experiments further show that stable aggregate accuracy conceals bidirectional question-level answer flips. The diagnostic also generalizes to four API-accessed models, whose VDG values range from 0.025 to 0.315. These results motivate VDG as a standard audit for whether video benchmarks measure visually grounded capability. Code is available at https://github.com/JaeLee18/accuracy-without-grounding.
Chinese Translation
视频大语言模型（LLMs）的基准准确性常被视为视觉理解的证据。我们对二十个模型进行了审计，这些模型的参数范围从2亿到78亿，涵盖十种架构类型。我们引入了视觉依赖差距（Visual Dependency Gap, VDG），即在原始视频和黑屏条件下每个问题的正确率差异。对MVBench进行的配对McNemar检验表明，准确性与视觉依赖是可分离的：模型在原始视频上的表现存在差异（p = 0.0003），而在黑屏上则没有差异（p = 0.53）。在各模型中，任务类型的排名是稳定的：属性感知（Attribute Perception）具有强烈的视觉特征，而时间推理（Temporal Reasoning）接近仅基于语言的基线。从黑屏到单帧、打乱帧和原始视频的诊断阶梯显示，帧的多样性提供了大部分视觉效益，而时间顺序在十六个开放权重模型中几乎不贡献准确性。将帧率从0.5调整到24 FPS的消融实验排除了稀疏采样作为原因。H.264实验进一步表明，稳定的总体准确性掩盖了双向问题级别答案的翻转。该诊断方法还可以推广到四个API访问的模型，其VDG值范围从0.025到0.315。这些结果促使VDG成为评估视频基准是否测量视觉基础能力的标准审计工具。代码可在 https://github.com/JaeLee18/accuracy-without-grounding 获取。
cs.CV / 14 / 2607.13318
Reflecting Process Expertise in Procedural Material Generation
在程序性材料生成中反映过程专业知识
Abstract
Procedural material creation underpins applications in digital content creation, visual effects, and 3D asset design. Achieving high-quality results requires more than reproducing node graphs -- it demands understanding the process by which experts construct materials. We formulate procedural material generation as retrieval-time process reasoning over expert demonstrations, elevating process to a first-class representation beyond graph-only synthesis. Concretely, we represent expert workflows as process traces: textual records of construction steps, parameters, and design intent. To instantiate this idea, we use a pretrained LLM-based ProcessSynthesizer to synthesize a process trace aligned with a user's intent and a pretrained LLM-based Compiler to ground the process trace into an executable Blender material graph. Because procedural expertise is most naturally conveyed through demonstrations, we leverage tutorial videos as a source of process knowledge and extract textual, LLM-compatible traces using automated video analysis tools. In an expert study with five Blender artists (avg. 7.5 years of experience), materials generated by reflecting expert demonstrations were found to produce workflows requiring fewer edits, and more closely match professional design strategies than methods operating solely on static artifacts. A user study with 150 participants further shows that our approach achieves superior generation and editing performance compared to prior procedural systems. All code, models, and data will be available at https://materialapprentice.github.io
Chinese Translation
程序性材料的创建是数字内容创作、视觉效果和3D资产设计中的基础。实现高质量的结果不仅仅需要重现节点图——它还要求理解专家构建材料的过程。我们将程序性材料生成形式化为对专家演示的检索时过程推理，将过程提升为超越仅图形合成的一类重要表示。具体而言，我们将专家工作流程表示为过程轨迹：构建步骤、参数和设计意图的文本记录。为了实现这一思想，我们使用基于预训练大语言模型（LLM）的ProcessSynthesizer合成与用户意图对齐的过程轨迹，并使用基于预训练LLM的Compiler将过程轨迹转化为可执行的Blender材料图。由于程序性专业知识最自然地通过演示传达，我们利用教程视频作为过程知识的来源，并使用自动视频分析工具提取文本的、兼容LLM的轨迹。在与五位Blender艺术家的专家研究中（平均7.5年经验），通过反映专家演示生成的材料被发现产生的工作流程需要更少的编辑，并且与专业设计策略的匹配程度高于仅基于静态工件的方法。对150名参与者的用户研究进一步表明，我们的方法在生成和编辑性能上优于先前的程序性系统。所有代码、模型和数据将可在 https://materialapprentice.github.io 获取。
cs.CV / 15 / 2607.13323
SARFA: Segment Anything with Radiomic Feature Alignment
SARFA：通过放射组学特征对齐进行任意分割
Abstract
The Segment Anything Model (SAM) has demonstrated strong generalizability across a variety of segmentation tasks. However, SAM often struggles in situations where the target to be segmented is ambiguous. This poses a problem in medical imaging, where accurate delineation of targets such as tumors is vital, but even expert radiologists can disagree on the appropriate boundary for a target. Addressing this, we propose SARFA (Segment Anything with Radiomic Feature Alignment), a novel framework for improved medical image segmentation. Via probabilistic prompting, SARFA generates a diverse set of plausible masks for each input image and optimizes them with a radiomics-driven training objective based on Fr\'echet Radiomic Distance (FRD) and Direct Preference Optimization (DPO). By minimizing the FRD between masked predicted and ground truth regions within each image, SARFA encourages segmentation outputs whose anatomical and textural characteristics align with clinically meaningful ground truth representations, without relying solely on pixel-level overlap. Evaluated on computed tomography (CT) and magnetic resonance imaging (MRI) benchmarks, SARFA outperforms existing ambiguous segmentation methods, demonstrating the effectiveness of radiomic feature alignment and DPO-style candidate mask ranking as a training objective. Our code is available at https://github.com/tbwa233/SARFA.
Chinese Translation
任意分割模型（Segment Anything Model, SAM）在多种分割任务中表现出了强大的泛化能力。然而，SAM在目标模糊的情况下往往表现不佳。这在医学成像中尤为突出，因为准确描绘肿瘤等目标的边界至关重要，但即使是专家放射科医生也可能对目标的适当边界存在分歧。为了解决这一问题，我们提出了SARFA（通过放射组学特征对齐进行任意分割），这是一个用于改善医学图像分割的新框架。通过概率提示，SARFA为每个输入图像生成一组多样化的可行掩码，并基于Fréchet放射组学距离（Fréchet Radiomic Distance, FRD）和直接偏好优化（Direct Preference Optimization, DPO）优化这些掩码。通过最小化每幅图像中掩码预测区域与真实区域之间的FRD，SARFA鼓励生成的分割输出在解剖和纹理特征上与临床意义明确的真实表示对齐，而不仅仅依赖于像素级重叠。在计算机断层扫描（CT）和磁共振成像（MRI）基准测试中，SARFA的表现优于现有的模糊分割方法，展示了放射组学特征对齐和DPO风格候选掩码排序作为训练目标的有效性。我们的代码可在 https://github.com/tbwa233/SARFA 获取。
cs.CV / 16 / 2607.13336
Delving into the Temporal Challenges of Unified Video Protection Against Image-to-Video and Fine-Tuning-based Customization
深入探讨统一视频保护在图像到视频和基于微调定制中的时间挑战
Abstract
Recent diffusion-based video generation models have enabled high-quality personalized video customization through both tuning-based pipelines, which fine-tune a video diffusion model, and reference-based pipelines such as image-to-video generation. However, these capabilities raise serious concerns about personal privacy, identity ownership and intellectual property protection. Existing anti-customization works focus on protecting images, while protection for videos against both reference- and tuning-based customization remains largely underexplored. Protecting videos in this setting raises three challenges: (i) Image-level perturbations, optimized frame by frame, cannot survive temporal compression by 3D video VAE. (ii) A video-level perturbation optimized on a single video is vulnerable to temporal editing and fails to protect unseen videos. (iii) Temporally inconsistent perturbations are not robust to temporal attacks. To address these challenges, we propose Temporally Consistent Universal Adversarial Perturbations (TC-UAP), the first protection method against both reference- and tuning-based video customization. TC-UAP optimizes an identity-level multi-frame UAP over sliding windows from multiple videos, accounting for local temporal dependencies induced by temporal compression in video VAE and enabling a single perturbation to protect unseen videos of varying lengths. Moreover, we introduce intrinsic temporal modeling and an extrinsic surrogate temporal-attack loss, which make the perturbation temporally consistent and robust to unseen temporal attacks. Empirically, quantitative and qualitative results show that TC-UAP achieves the strongest identity protection compared with existing methods under both reference- and tuning-based video customization, and remains robust under multiple unseen temporal attacks.
Chinese Translation
近期基于扩散的视频生成模型通过调优管道（对视频扩散模型进行微调）和基于参考的管道（如图像到视频生成）实现了高质量的个性化视频定制。然而，这些能力引发了关于个人隐私、身份所有权和知识产权保护的严重担忧。现有的反定制工作主要集中在保护图像，而针对视频的参考和调优定制的保护仍然在很大程度上未被探索。在这种情况下保护视频面临三个挑战：（i）逐帧优化的图像级扰动无法在3D视频变分自编码器（VAE）中抵御时间压缩。（ii）在单个视频上优化的视频级扰动容易受到时间编辑的影响，无法保护未见过的视频。（iii）时间不一致的扰动对时间攻击不具备鲁棒性。为了解决这些挑战，我们提出了时间一致的通用对抗扰动（Temporally Consistent Universal Adversarial Perturbations, TC-UAP），这是首个针对参考和基于微调的视频定制的保护方法。TC-UAP在多个视频的滑动窗口上优化身份级多帧UAP，考虑了视频VAE中时间压缩引起的局部时间依赖性，使得单一扰动能够保护不同长度的未见视频。此外，我们引入了内在时间建模和外在替代时间攻击损失，使得扰动在时间上保持一致并对未见的时间攻击具有鲁棒性。实证结果表明，TC-UAP在参考和基于微调的视频定制下相比现有方法实现了最强的身份保护，并在多种未见的时间攻击下保持鲁棒性。
cs.CV / 17 / 2607.13343
Marker-free deformable registration and fusion for augmented reality-guided positive margin localization during tumor resection surgery
无标记变形配准与融合用于增强现实引导的肿瘤切除手术中的阳性边缘定位
Abstract
Positive margins in head and neck oncologic surgery require mapping specimen-side pathology findings to the patient resection bed. This is challenging because pathologists identify the positive margin on slices of the resected, deformed specimen, while surgeons must relocate the corresponding site on the resection bed using only verbal descriptions and no visual guidance. We present a marker-free augmented reality (AR) workflow for mapping a margin label from a three-dimensional specimen scan to the resection bed. The method combines contour-constrained deformation, residual alignment to a depth scan, surface-based fusion to a head-mounted display, and target projection onto the reconstructed bed. Bead-suture correspondences estimate specimen deformation, whereas patient-to-display fusion does not require external fiducial markers. Following formative experiments, five residents and surgeons performed cadaveric cheek and scalp re-resection tasks under verbal guidance, verbal guidance with specimen examination, and AR guidance. Deformation target errors were $7.63 \pm 3.74$ mm for the cheek and $3.72 \pm 1.02$ mm for the scalp; residual specimen-to-bed distances were $2.43 \pm 2.15$ mm and $2.19 \pm 1.06$ mm, respectively. Fusion error did not differ significantly between marker-free and marker-based methods on either cadaver; overall marker-free fusion error was $2.15 \pm 0.87$ mm. End-to-end margin localization error decreased from $21.40 \pm 3.84$ mm with verbal guidance and $16.09 \pm 4.30$ mm with specimen examination to $6.19 \pm 1.79$ mm with AR guidance ($p < 0.001$). Online fusion required $5.23 \pm 0.34$ s. These results demonstrate effective marker-free AR guidance for positive-margin localization and support more precise tumor resection.
Chinese Translation
在头颈部肿瘤外科手术中，阳性边缘需要将标本侧的病理发现映射到患者的切除床上。这一过程具有挑战性，因为病理学家在切除的变形标本的切片上识别阳性边缘，而外科医生则必须仅通过口头描述而没有视觉指导来重新定位切除床上的相应位置。我们提出了一种无标记的增强现实（AR）工作流程，将边缘标签从三维标本扫描映射到切除床。该方法结合了轮廓约束变形、与深度扫描的残余对齐、基于表面的融合到头戴显示器以及在重建床上的目标投影。珠子-缝合对应关系用于估计标本变形，而患者与显示器的融合不需要外部基准标记。经过初步实验，五名住院医生和外科医生在口头指导、口头指导结合标本检查以及AR指导下执行了尸体面颊和头皮的再切除任务。面颊的变形目标误差为 $7.63 imes 3.74$ mm，头皮为 $3.72 imes 1.02$ mm；残余标本与切除床的距离分别为 $2.43 imes 2.15$ mm 和 $2.19 imes 1.06$ mm。无标记和有标记方法在任一尸体上的融合误差没有显著差异；整体无标记融合误差为 $2.15 imes 0.87$ mm。端到端的边缘定位误差从口头指导下的 $21.40 imes 3.84$ mm 和标本检查下的 $16.09 imes 4.30$ mm 降低到 AR 指导下的 $6.19 imes 1.79$ mm（$p < 0.001$）。在线融合所需时间为 $5.23 imes 0.34$ s。这些结果表明无标记的 AR 指导在阳性边缘定位中有效，并支持更精确的肿瘤切除。
cs.CV / 18 / 2607.13345
Audio-Text Cross-Attention with Psycholinguistic Support Features for Ambivalence/Hesitancy Recognition
基于心理语言学支持特征的音频-文本交叉注意力模型用于矛盾/犹豫识别
Abstract
We present an audio-text system for the Ambivalence/Hesitancy Video Recognition Challenge of the 11th ABAW Competition. The method excludes visual frames and represents each video as overlapping 5-second windows aligned with transcript timestamps. Each window combines a 320-dimensional prosodic audio descriptor, a 768-dimensional emotion-oriented RoBERTa embedding, and 74 handcrafted features capturing uncertainty, hedging, and attitudinal conflict. Audio and text are fused via temporal cross-attention, while support features are injected prior to gated multiple-instance learning (MIL) pooling to modulate the window's importance. Predictions from five independently initialized models are averaged. On the labeled public development set, the ensemble achieved an average precision of 0.875 and a macro-F1 of 0.72. Our source code is publicly available at https://github.com/Liga-de-IA-PUCPR/abaw-11-ah-challenge/.
Chinese Translation
我们提出了一种音频-文本系统，用于第11届ABAW比赛的矛盾/犹豫视频识别挑战。该方法排除了视觉帧，将每个视频表示为与转录时间戳对齐的重叠5秒窗口。每个窗口结合了一个320维的韵律音频描述符、一个768维的情感导向RoBERTa嵌入以及74个手工制作的特征，这些特征捕捉了不确定性、模棱两可和态度冲突。音频和文本通过时间交叉注意力进行融合，而支持特征则在门控多实例学习（MIL）池化之前注入，以调节窗口的重要性。来自五个独立初始化模型的预测结果进行平均。在标记的公共开发集上，该集成模型实现了0.875的平均精度和0.72的宏F1分数。我们的源代码可在https://github.com/Liga-de-IA-PUCPR/abaw-11-ah-challenge/获取。
cs.CV / 19 / 2607.13361
Detector Confidence Signals Presence Rather Than Occlusion in Cluttered Manipulation
检测器置信度信号指示存在而非遮挡于杂乱的操作中
Abstract
Occlude a named object until about an eighth of it remains visible, and an open-vocabulary detector's confidence that the object is present barely changes; as the clutter around it grows the confidence can even rise. On real video the detector still reports the object present in 99% of occluded frames, on another instance of the same category. This matters because that confidence is widely read as a visibility signal, used to threshold detections, evaluate open-vocabulary detectors, ground language, retrieve instances, and gate active perception. We audit whether it reflects occlusion by pairing every view with a geometry-segmentation oracle that gives detector-free ground-truth visibility. As true visibility falls from every scene to one in eight, the confidence stays nearly constant and uncorrelated with visibility, and the detector reports the target present in about nine of ten scenes, firing on same-category distractors: it signals that the category is present somewhere, not that the specific target is visible. The failure holds across three detectors (Grounding DINO, OWLv2, and Segment Anything Model 3), nine object categories, two simulators with different renderers and object sets, built and natural occlusion, and real video. Two consequences follow: a confidence-based metric understates the value of resolving occlusion by about ten times (8 against 88 points in our active-perception setting), and a confidence-based gate fires exactly when the object is hidden. No single-view signal we tried, including a realizable localization check, flags the occlusion, because the occluders sit where the target is. We connect the effect to detector miscalibration and object hallucination, release the controlled benchmark, and recommend target-grounded signals for gating and evaluation.
Chinese Translation
将一个命名对象遮挡，直到大约八分之一的部分可见，开放词汇检测器对该对象存在的置信度几乎没有变化；随着周围杂物的增加，置信度甚至可能上升。在真实视频中，检测器仍然在99%的遮挡帧中报告该对象存在，且为同一类别的另一个实例。这一点很重要，因为该置信度通常被视为可见性信号，用于阈值检测、评估开放词汇检测器、基础语言、检索实例和激活感知。我们通过将每个视图与几何分割神谕配对来审计其是否反映遮挡，该神谕提供无检测器的真实可见性。随着真实可见性从每个场景下降到八分之一，置信度几乎保持不变，并且与可见性无关，检测器在大约十分之九的场景中报告目标存在，并对同类别的干扰物发出信号：它表明该类别在某处存在，而不是特定目标可见。此失败在三种检测器（Grounding DINO、OWLv2 和 Segment Anything Model 3）、九个物体类别、两个具有不同渲染器和物体集的模拟器、构建和自然遮挡以及真实视频中均存在。由此产生两个后果：基于置信度的度量低估了解决遮挡的价值约十倍（在我们的主动感知设置中为8对88分），而基于置信度的门控在对象被隐藏时恰好触发。我们尝试的没有任何单视图信号，包括可实现的定位检查，能够标识遮挡，因为遮挡物位于目标所在的位置。我们将这一效应与检测器的误校准和对象幻觉联系起来，发布了受控基准，并建议使用目标基础信号进行门控和评估。
cs.CV / 20 / 2607.13365
DiffGI: Differentiable Geometry Images for High-Fidelity Thin-Shell 3D Generation
DiffGI：用于高保真薄壳3D生成的可微分几何图像
Abstract
Existing 3D generative models predominantly rely on implicit volumetric representations, which enforce watertight topology and struggle to represent thin-shell and non-manifold geometries such as garments. Geometry image-based approaches offer a surface-centric alternative, but existing methods rely on discrete binary occupancy maps whose resolution-dependent boundary encoding causes staircase artifacts and information loss upon downsampling, while surface reconstruction remains a non-differentiable post-processing step disconnected from the learning pipeline. To address this, we propose Differentiable Geometry Image (DiffGI), an end-to-end 3D-to-2D mapping framework that seamlessly integrates surface representation and geometric optimization. DiffGI replaces binary maps with a continuous 2D Truncated Signed Distance Function (TSDF), which encodes boundary position at subpixel precision within a fixed grid resolution, eliminating resolution-dependent staircase artifacts even under aggressive downsampling. Building on this continuous field, we introduce a differentiable Marching Squares algorithm based on analytical linear interpolation, allowing gradients from 3D surface losses to propagate back to the 2D latent space. Leveraging this differentiable pipeline, we train a DiffGI-VAE augmented with a geometry-aware normal rendering loss to compress complex 3D surfaces into an ultra-compact 32X32 latent space, and instantiate a transformer-based latent diffusion model with a flow-matching objective on top of this space for conditional 3D generation. Extensive experiments on garment and object datasets demonstrate that our method achieves superior reconstruction fidelity and boundary precision compared to prior geometry-image and voxel-based approaches, while requiring significantly fewer computational resources.
Chinese Translation
现有的3D生成模型主要依赖于隐式体积表示，这种表示强制要求水密拓扑，并且难以表示薄壳和非流形几何体，例如服装。基于几何图像的方法提供了一种以表面为中心的替代方案，但现有方法依赖于离散的二进制占用图，其分辨率依赖的边界编码在下采样时会导致阶梯伪影和信息损失，而表面重建仍然是一个与学习管道无关的不可微分后处理步骤。为了解决这个问题，我们提出了可微分几何图像（Differentiable Geometry Image，DiffGI），这是一个端到端的3D到2D映射框架，能够无缝集成表面表示和几何优化。DiffGI用连续的二维截断符号距离函数（Truncated Signed Distance Function，TSDF）替代二进制图，这种方法在固定网格分辨率内以亚像素精度编码边界位置，消除了即使在激进下采样下也会出现的分辨率依赖的阶梯伪影。在此连续场的基础上，我们引入了一种基于解析线性插值的可微分行进方格算法，允许来自3D表面损失的梯度向2D潜在空间传播。利用这一可微分管道，我们训练了一个DiffGI-VAE，并增强了几何感知的法线渲染损失，以将复杂的3D表面压缩到超紧凑的32X32潜在空间，并在此空间上实例化了一个基于变换器的潜在扩散模型，采用流匹配目标进行条件3D生成。在服装和物体数据集上的大量实验表明，我们的方法在重建保真度和边界精度方面优于先前的几何图像和体素基础方法，同时所需的计算资源显著减少。
cs.CV / 21 / 2607.13371
RoughNet: Mapping Arctic Sea Ice Roughness Using Diffusion-Based Super-Resolution of Satellite Imagery
RoughNet：基于扩散的卫星图像超分辨率映射北极海冰粗糙度
Abstract
Accurate estimation of landfast sea ice roughness is critical for climate modeling and safe Arctic over-ice travel, yet existing approaches rely on costly airborne surveys or sparse in-situ measurements, limiting spatial coverage and operational scalability. Here we show that high-resolution sea ice topography can be reconstructed directly from optical satellite imagery using a conditional diffusion framework. Our approach, RoughNet, learns to map 10 m Sentinel-2 multispectral images to locally normalized 1 m surface elevation residual fields, enabling fine-scale roughness characterization from widely available satellite data. Trained on airborne LiDAR data from two Arctic regions and evaluated on an unseen third Arctic region, the model generalizes across diverse ice conditions and partially reproduces small-scale topographic structure. The best-performing model achieves an out-of-domain root mean squared error of 9 cm while preserving the statistical and spectral properties of the underlying roughness field. These results demonstrate that generative diffusion models can recover physically meaningful surface structure from optical imagery alone, providing a scalable pathway for high-resolution sea ice mapping and roughness estimation in data-sparse environments.
Chinese Translation
准确估计陆地固定海冰的粗糙度对于气候建模和安全的北极冰面旅行至关重要，然而现有方法依赖于昂贵的空中调查或稀疏的现场测量，限制了空间覆盖范围和操作可扩展性。在此，我们展示了如何利用条件扩散框架直接从光学卫星图像重建高分辨率海冰地形。我们的方法RoughNet学习将10米的Sentinel-2多光谱图像映射到局部归一化的1米表面高程残差场，从而能够从广泛可用的卫星数据中进行细尺度的粗糙度特征化。该模型在来自两个北极地区的空中激光雷达数据上进行训练，并在一个未见过的第三个北极地区进行评估，能够在多样的冰况下进行泛化，并部分再现小尺度地形结构。表现最佳的模型在域外的均方根误差为9厘米，同时保持了基础粗糙度场的统计和光谱特性。这些结果表明，生成性扩散模型能够仅通过光学图像恢复物理上有意义的表面结构，为在数据稀缺环境中进行高分辨率海冰映射和粗糙度估计提供了一条可扩展的途径。
cs.CV / 22 / 2607.13386
FM$^2$: Unified Federated Foundation Models for Heterogeneous Multimodal Medical Imaging
FM$^2$: 异构多模态医学影像的统一联邦基础模型
Abstract
Building foundation models for medical imaging requires pooling data across institutions, yet privacy regulations prohibit centralized aggregation. Existing Federated Foundation Models either fine-tune natural-image models with poor medical-domain transfer, or train from scratch within a single modality, lacking the flexibility to unify tasks. We identify an under-explored challenge, Imaging Modality Heterogeneity, where clients operate under two structural regimes: Overlapped (shared modalities with heterogeneous label distributions) and Non-overlapped (fully disjoint modalities per client). We propose FM$^2$, a unified framework that trains the core backbone from scratch to preserve medical domain fidelity while optionally incorporating biomedical pretrained encoders for vision-language alignment. FM$^2$ equips each client with dual Mixture-of-Experts modules (a Class-wise MoE for personalized category knowledge and a Domain-wise MoE for shared cross-modality representations), coupled with a Heterogeneous Modality Alignment (HMA) regularizer that explicitly aligns modality-specific expert parameters, admitting provable $O(1/\sqrt{T})$ convergence and generalization guarantees. FM$^2$ further incorporates Caption-Enhanced Learning (CEL), where locally retained GPT-4o-generated captions serve as a textual semantic bridge enabling representation transfer across clients with disjoint modalities, and demonstrates extensibility to Federated Medical VQA. Experiments on our MIMH benchmark (classification and CEL) and real-world medical VQA datasets confirm consistent superiority over state-of-the-art federated baselines and strong out-of-modality generalization across all three tasks.
Chinese Translation
构建医学影像的基础模型需要跨机构汇聚数据，但隐私法规禁止集中聚合。现有的联邦基础模型要么对自然图像模型进行微调，但在医学领域的迁移效果较差，要么在单一模态内从头训练，缺乏统一任务的灵活性。我们识别出一个尚未充分探索的挑战：影像模态异质性，其中客户端在两种结构模式下运行：重叠（共享模态但标签分布异质）和非重叠（每个客户端完全独立的模态）。我们提出FM$^2$，一个统一框架，从头开始训练核心骨干，以保持医学领域的保真度，同时可选地结合生物医学预训练编码器以实现视觉-语言对齐。FM$^2$为每个客户端配备双重专家混合模块（一个用于个性化类别知识的类别级MoE和一个用于共享跨模态表示的领域级MoE），并结合异构模态对齐（HMA）正则化器，明确对齐模态特定的专家参数，确保可证明的$O(1/ ext{sqrt}{T})$收敛和泛化保证。FM$^2$进一步结合了增强标题学习（CEL），其中本地保留的GPT-4o生成的标题作为文本语义桥梁，促进在模态不重叠的客户端之间的表示转移，并展示了对联邦医学视觉问答（VQA）的扩展性。在我们的MIMH基准（分类和CEL）及真实世界医学VQA数据集上的实验确认了其在所有三项任务中相较于最先进的联邦基线的一致优越性及强大的跨模态泛化能力。
cs.CV / 23 / 2607.13409
AnomExpert: Identifying and Selecting Anatomical Planes for Prenatal Ultrasound Anomaly Diagnosis
AnomExpert：识别和选择产前超声异常诊断的解剖平面
Abstract
Life-limiting congenital anomalies require accurate prenatal diagnosis for appropriate clinical decision-making. Prenatal ultrasound (US) examinations involve multiple anatomical planes, and diagnosis depends on identifying anatomical planes and selecting diagnostically relevant planes for each anomaly. Existing automated methods either rely on plane-level annotations or aggregate heterogeneous images without explicitly modeling these diagnostic capabilities. We propose AnomExpert, a prototype-driven framework for prenatal US anomaly diagnosis using only case-level supervision. AnomExpert introduces learnable plane prototypes to organize unordered images into latent representations corresponding to anatomical planes without requiring plane annotations. A disease-aware sparse selection mechanism further selects diagnostically relevant planes for each anomaly. Experiments on a multi-center dataset of 3,654 cases show that AnomExpert consistently outperforms nine representative multi-instance learning methods. Using a ViT-small backbone, it achieves 86.9% accuracy and 84.2% F1-score while maintaining parameter efficiency. These findings indicate that modeling anatomical plane identification and disease-specific plane selection improves weakly supervised multi-plane prenatal US anomaly classification. The code is available at https://github.com/TIanCat/AnomExpert.
Chinese Translation
生命限制性先天性异常需要准确的产前诊断以便进行适当的临床决策。产前超声（US）检查涉及多个解剖平面，诊断依赖于识别解剖平面并为每个异常选择具有诊断相关性的平面。现有的自动化方法要么依赖于平面级注释，要么聚合异构图像而未明确建模这些诊断能力。我们提出了AnomExpert，一个基于原型的框架，用于仅使用案例级监督的产前US异常诊断。AnomExpert引入可学习的平面原型，将无序图像组织成对应于解剖平面的潜在表示，而无需平面注释。一个疾病感知的稀疏选择机制进一步为每个异常选择具有诊断相关性的平面。在一个包含3,654个案例的多中心数据集上的实验表明，AnomExpert始终优于九种代表性的多实例学习方法。使用ViT-small骨干网络，它实现了86.9%的准确率和84.2%的F1-score，同时保持参数效率。这些发现表明，建模解剖平面识别和疾病特异性平面选择改善了弱监督的多平面产前US异常分类。代码可在 https://github.com/TIanCat/AnomExpert 获取。
cs.CV / 24 / 2607.13415
MultiAnimate: A Unified Framework for Controllable Multi-Character Animation
MultiAnimate：可控多角色动画的统一框架
Abstract
Recent advances in generative models and technological innovations have significantly addressed the fundamental challenges of character image animation. However, existing approaches predominantly focus on character animation from a single reference image, substantially limiting their applicability in scenarios such as multiple character interaction animation. To fill this gap, this paper introduces MultiAnimate, a comprehensive framework that enables concurrent animation of multiple characters within a shared environment while preserving both identity consistency and spatial relationships. The framework achieves these objectives through multiple well-designed mechanisms. First, we incorporate an identity-specific reference net that enables appearance extraction from multiple reference images, distinguishing MultiAnimate from existing approaches constrained to single reference inputs. Second, we implement an identity-aware pose encoder to address the character-pose binding challenge, wherein an attention mechanism enables the network to accurately differentiate and process multiple pose sequences during generation. Third, we introduce an interaction guider module that enhances the framework's capability to handle complex inter-character interactions by leveraging character-specific mask information, serving as an optional component that refines the pose sequences. Extensive experiments and ablation analyses demonstrate our framework's superiority in multiple character animation, particularly in scenarios involving complex motion sequences.
Chinese Translation
近期生成模型和技术创新的进展显著解决了角色图像动画的基本挑战。然而，现有方法主要集中于从单一参考图像进行角色动画，这在多个角色交互动画等场景中的适用性受到限制。为填补这一空白，本文提出了MultiAnimate，一个全面的框架，能够在共享环境中同时对多个角色进行动画，同时保持身份一致性和空间关系。该框架通过多个精心设计的机制实现这些目标。首先，我们引入了一个身份特定的参考网络，能够从多个参考图像中提取外观，使MultiAnimate区别于现有仅限于单一参考输入的方法。其次，我们实现了一个身份感知的姿态编码器，以解决角色与姿态绑定的挑战，其中一个注意力机制使网络能够在生成过程中准确区分和处理多个姿态序列。第三，我们引入了一个交互引导模块，通过利用角色特定的掩码信息，增强了框架处理复杂角色间交互的能力，作为一个可选组件来优化姿态序列。大量实验和消融分析表明，我们的框架在多个角色动画方面的优越性，尤其是在涉及复杂运动序列的场景中。
cs.CV / 25 / 2607.13421
ScanFocus: A Coarse-to-Fine Framework for Spatio-Temporal Video Grounding
ScanFocus：一种粗到细的时空视频定位框架
Abstract
Spatio-Temporal Video Grounding (STVG) aims to retrieve the visual trajectory of a specific object from a video stream as described by a natural language expression. However, most advanced methods struggle to balance global context modeling with precise boundary localization. Due to the prohibitive computational costs of processing long videos, these approaches typically resort to low-rate temporal downsampling and implicit motion modeling. This inevitably suppresses high-frequency boundary cues and neglects the explicit inter-frame dependencies required for precise boundary delineation. To address these limitations, we present \textbf{ScanFocus}, a novel coarse-to-fine framework that decouples the STVG task into a global spatio-temporal scan and a local boundary focus. Specifically, we utilize a unified vision-language fusion encoder combined with a lightweight Deformable Semantic-Motion Fusion module to efficiently align multimodal features and generate coarse proposals. To recover the suppressed fine-grained details, we introduce the Semantic-Guided Temporal Aggregator (SGTA) in the refinement stage. By densely sampling around coarse boundaries, SGTA explicitly models short-term temporal interactions under semantic guidance, capturing rapid motion changes for precise timestamp regression. Extensive experiments on three widely used benchmarks demonstrate the performance superiority of our proposed method over previous approaches. Code will be released at https://github.com/TenMinutes209/ScanFocus.
Chinese Translation
时空视频定位（STVG）旨在从视频流中检索特定对象的视觉轨迹，该轨迹由自然语言表达描述。然而，大多数先进的方法在全球上下文建模与精确边界定位之间难以取得平衡。由于处理长视频的计算成本过高，这些方法通常采用低频率的时间下采样和隐式运动建模。这不可避免地抑制了高频边界线索，并忽视了精确边界描绘所需的显式帧间依赖关系。为了解决这些限制，我们提出了 extbf{ScanFocus}，一种新颖的粗到细框架，将STVG任务解耦为全局时空扫描和局部边界聚焦。具体而言，我们利用统一的视觉-语言融合编码器，并结合轻量级的可变形语义-运动融合模块，以高效对齐多模态特征并生成粗略提案。为了恢复被抑制的细粒度细节，我们在细化阶段引入了语义引导时间聚合器（SGTA）。通过在粗略边界周围进行密集采样，SGTA在语义引导下显式建模短期时间交互，捕捉快速运动变化以实现精确的时间戳回归。在三个广泛使用的基准测试上的大量实验表明，我们提出的方法在性能上优于之前的方法。代码将发布在 https://github.com/TenMinutes209/ScanFocus。
cs.CV / 26 / 2607.13437
CLIP-Guided Label-Free Discriminative Region Scoring for Fine-Grained Classification
基于CLIP引导的无标签细粒度分类判别区域评分
Abstract
Recent vision models such as CLIP and SAM enable training-free segmentation and semantic encoding for fine-grained classification. A common approach is to compare the representations of segmented image regions with the text prompt embeddings of the corresponding labels. However, it remains unclear how different local regions and CLIP-based scoring strategies affect the selection of discriminative evidence, especially when ground-truth labels are unavailable. In this paper, we propose a unified CLIP-guided label-free region scoring framework for fine-grained classification. The framework evaluates cosine similarity-based, margin-based, and entropy-based scoring strategies using both SAM-generated masks and random crops, and introduces two label-free pseudo-label variants based on global image embeddings and local region embeddings. We conduct experiments on five fine-grained classification datasets to systematically compare different region generation methods and scoring strategies. The results show that Soft Negative Margin scoring achieves the strongest performance, and pseudo-label scoring closely approximates true-label performance. Although SAM produces semantically meaningful masks, random-crop-based pseudo-label scoring consistently outperforms SAM-based scoring across all datasets, suggesting that random crops preserve surrounding information and provide more stable semantic context when pseudo-labels are noisy. In addition, SAM masks benefit from aggregating embeddings from all regions, whereas random crops tend to perform better with a smaller top-k subset. These findings provide new insights for fine-grained classification.
Chinese Translation
近期的视觉模型如CLIP和SAM使得无训练的分割和语义编码成为可能，从而应用于细粒度分类。一种常见的方法是将分割图像区域的表示与相应标签的文本提示嵌入进行比较。然而，不同局部区域和基于CLIP的评分策略如何影响判别证据的选择仍不清楚，尤其是在缺乏真实标签的情况下。本文提出了一种统一的基于CLIP引导的无标签区域评分框架，用于细粒度分类。该框架评估基于余弦相似度、边际和熵的评分策略，使用SAM生成的掩膜和随机裁剪，并引入了两种基于全局图像嵌入和局部区域嵌入的无标签伪标签变体。我们在五个细粒度分类数据集上进行了实验，以系统比较不同的区域生成方法和评分策略。结果表明，Soft Negative Margin评分实现了最佳性能，伪标签评分与真实标签性能接近。尽管SAM生成了语义上有意义的掩膜，但基于随机裁剪的伪标签评分在所有数据集中始终优于基于SAM的评分，这表明随机裁剪保留了周围信息，并在伪标签噪声较大时提供了更稳定的语义上下文。此外，SAM掩膜通过聚合所有区域的嵌入受益，而随机裁剪在较小的top-k子集上表现更佳。这些发现为细粒度分类提供了新的见解。
cs.CV / 27 / 2607.13449
DreamSat-Pose: Spacecraft Pose Estimation from Single-View 3D Reconstructions and Learned 2D-3D Feature Matching
DreamSat-Pose：基于单视图3D重建和学习的2D-3D特征匹配的航天器姿态估计
Abstract
6-DoF pose estimation is a critical task in autonomous rendezvous and proximity operations. In the case of an unknown target, this task becomes challenging as it shall be paired with the reconstruction of the target shape model. In this article, we propose a novel framework for single-shot shape and pose estimation of unknown spacecraft objects. Given a single image, we first reconstruct a 3D shape model of the target, then estimate the relative six-degrees-of-freedom pose by learning dense 2D-3D correspondences. The image features are extracted using a frozen DINOv3 vision transformer, while the geometric features are computed from the reconstructed point cloud using a trainable dynamic graph convolutional neural network encoder. A dual-stream transformer matcher refines descriptors through alternating self- and cross-attention, producing soft correspondences that are passed to a Perspective-$n$-Point solver for pose recovery. We evaluate the method on the SPE3R dataset and consider FoundationPose as a representative baseline for current state-of-the-art capabilities. Results show reliable pose estimates achieving 0.157 degrees mean pointing error using only a single image and reconstructed geometry, demonstrating strong generalization to unseen spacecraft.
Chinese Translation
六自由度（6-DoF）姿态估计是自主交会与近距离操作中的一项关键任务。在目标未知的情况下，这项任务变得更加具有挑战性，因为它需要与目标形状模型的重建相结合。本文提出了一种新颖的框架，用于对未知航天器对象进行单次形状和姿态估计。给定一幅单一图像，我们首先重建目标的3D形状模型，然后通过学习稠密的2D-3D对应关系来估计相对的六自由度姿态。图像特征通过冻结的DINOv3视觉变换器提取，而几何特征则通过可训练的动态图卷积神经网络编码器从重建的点云中计算得出。双流变换器匹配器通过交替的自注意力和交叉注意力来优化描述符，生成软对应关系，这些关系随后传递给透视-$n$-点求解器以恢复姿态。我们在SPE3R数据集上评估该方法，并将FoundationPose作为当前最先进能力的代表基线。结果表明，仅使用单幅图像和重建几何体就能实现可靠的姿态估计，平均指向误差达到0.157度，展示了对未见航天器的强泛化能力。
cs.CV / 28 / 2607.13452
Symbiosis-Inspired Knowledge Distillation for Incremental Object Detection
基于共生启发的增量目标检测知识蒸馏
Abstract
Incremental object detection (IOD) aims to extend detectors to new categories while retaining previously acquired knowledge. Existing methods often adopt a class incremental learning perspective, separating feature spaces to sharpen decision boundaries. However, this separation-oriented paradigm may overlook object symbiosis in detection, where co-occurrence and occlusion introduce spatial and semantic dependencies that benefit from shared representations. Ignoring these dependencies distorts the shared representations, exacerbates confusion between old and new classes, and accelerates catastrophic forgetting. To address this, we propose Symbiosis-Inspired Knowledge Distillation (SIKD), which explicitly leverages object symbiosis at two complementary levels. Spatial Symbiosis Distillation (SpSD) focuses on symbiotic regions where the old model responds with high overlap to objects in the new task. It preserves generalizable old class cues, suppresses class-specific bias and redundancy, and distills the refined evidence to the new model at matched spatial locations with slot-aligned supervision. Semantic Symbiosis Distillation (SeSD) maintains class level structure by forming confidence weighted prototypes for old classes and aligning their inter class soft ranks over the old class logits, which stabilizes the semantic topology during adaptation. Extensive experiments demonstrate the effectiveness and superiority of the proposed method.
Chinese Translation
增量目标检测（IOD）旨在在保留先前获得知识的同时，将检测器扩展到新类别。现有方法通常采用类增量学习的视角，分离特征空间以锐化决策边界。然而，这种以分离为导向的范式可能忽视了检测中的对象共生现象，其中共现和遮挡引入了空间和语义依赖，这些依赖受益于共享表示。忽视这些依赖会扭曲共享表示，加剧旧类和新类之间的混淆，并加速灾难性遗忘。为了解决这个问题，我们提出了基于共生启发的知识蒸馏（SIKD），该方法在两个互补层面上明确利用对象共生。空间共生蒸馏（SpSD）专注于共生区域，在这些区域中，旧模型对新任务中的对象响应高度重叠。它保留了可泛化的旧类线索，抑制了类特定的偏差和冗余，并在匹配的空间位置通过槽对齐的监督将精炼证据蒸馏到新模型中。语义共生蒸馏（SeSD）通过为旧类形成置信加权原型并对旧类逻辑回归的类间软排名进行对齐，维持类级结构，从而在适应过程中稳定语义拓扑。大量实验表明，所提方法的有效性和优越性。
cs.CV / 29 / 2607.13454
GeoAnchor: Collaborative Reasoning via Latent Decomposition for 3D Spatial Understanding
GeoAnchor：通过潜在分解进行协同推理以实现3D空间理解
Abstract
Although multimodal large language models (MLLMs) have achieved remarkable progress, understanding 3D spatial relationships from 2D images remains a critical challenge. Existing methods primarily rely on symbolic text tokens, which inherently lack the fidelity to represent continuous geometric information. While recent methods use latent representations to enhance reasoning, relying on a single latent type cannot adapt to the diversity of spatial tasks, leading to misalignment in complex geometric scenarios. To address these limitations, we propose GeoAnchor, an interleaved text-latent reasoning framework. GeoAnchor decomposes 3D spatial information into three complementary components: position latents for object grounding, direction latents for relational orientation, and geometry latents for scene structure. These components are recombined in a structured space to construct local evidence while capturing global context, enabling dynamic and interpretable reasoning. Furthermore, we introduce a collaborative training strategy that guides the model from local spatial perception to comprehensive 3D understanding. Extensive experiments on diverse and complex 3D reasoning tasks demonstrate that GeoAnchor outperforms the state of the art, validating its effectiveness and generalization capabilities.
Chinese Translation
尽管多模态大型语言模型（MLLMs）已取得显著进展，但从2D图像中理解3D空间关系仍然是一个关键挑战。现有方法主要依赖符号文本标记，这本质上缺乏表示连续几何信息的精确性。虽然近期方法使用潜在表示来增强推理，但依赖单一潜在类型无法适应空间任务的多样性，导致在复杂几何场景中的错位。为了解决这些局限性，我们提出了GeoAnchor，一个交错的文本-潜在推理框架。GeoAnchor将3D空间信息分解为三个互补组件：用于物体定位的位置信息潜在（position latents）、用于关系定向的方向信息潜在（direction latents）和用于场景结构的几何信息潜在（geometry latents）。这些组件在结构化空间中重新组合，以构建局部证据，同时捕捉全局上下文，从而实现动态和可解释的推理。此外，我们引入了一种协同训练策略，引导模型从局部空间感知到全面的3D理解。在多样化和复杂的3D推理任务上的广泛实验表明，GeoAnchor的表现超越了现有最先进技术，验证了其有效性和泛化能力。
cs.CV / 30 / 2607.13456
TreeSRNF: Square-Root Normal Fields for Generative Modelling of the Geometric and Structural Variability in Tree-like 3D Objects
TreeSRNF：用于生成树状三维物体几何和结构变异性的平方根法线场
Abstract
We introduce a novel mathematical framework for analyzing and generating complex tree-shaped 3D objects, such as botanical trees and plants, which deform both in their 3D geometry and branching structure. Unlike previous works, which either consider only the skeletal structure of tree-like objects or approximate their 3D geometry using branch thickness, the proposed framework accurately models both the 3D geometry of the tree branches and the way they are interconnected. In this paper, we first generalize the Square Root Normal Fields (SRNF) representation, originally proposed for the statistical analysis of genus-0 surfaces, to tree-shaped 3D objects. We then treat tree-shaped 3D objects as points on a novel Riemannian tree-shape space equipped with a novel Riemannian metric that measures the amount of surface bending and stretching, and structural changes one needs to apply to one 3D tree-shape to align it with another. This way, deformations become trajectories in this novel tree-shape space. We analyze the theoretical properties of this novel tree-shape space and the corresponding metric and develop algorithms for computing point-wise and branch-wise correspondences and geodesic paths between complex 3D trees. We finally show how to use these building blocks for (1) computing statistical summaries, \ie means and modes of variation, of collections of tree-shaped 3D objects, and (2) synthesizing novel tree-shaped 3D objects by sampling from probability distributions fitted to a population of tree-shaped 3D objects. We demonstrate the performance and utility of the proposed framework on real and synthetic plants and botanical trees and show that it significantly outperforms the state-of-the-art.
Chinese Translation
我们提出了一种新颖的数学框架，用于分析和生成复杂的树状三维物体，例如植物树木和植物，这些物体在其三维几何形状和分支结构上均会发生变形。与之前的研究不同，后者要么仅考虑树状物体的骨架结构，要么通过分支厚度来近似其三维几何形状，而我们提出的框架准确地建模了树枝的三维几何形状及其相互连接的方式。本文首先将原本用于零 genus 表面的统计分析的平方根法线场（Square Root Normal Fields, SRNF）表示推广到树状三维物体。然后，我们将树状三维物体视为具有新颖黎曼树形空间中的点，该空间配备了一种新颖的黎曼度量，用于测量将一个三维树形与另一个对齐所需施加的表面弯曲、拉伸和结构变化的量。通过这种方式，变形成为这个新颖树形空间中的轨迹。我们分析了这个新颖树形空间及其相应度量的理论性质，并开发了计算复杂三维树木之间点对点和分支间对应关系及测地路径的算法。最后，我们展示了如何利用这些构建模块（1）计算树状三维物体集合的统计摘要，即均值和变异模式，以及（2）通过从拟合于树状三维物体群体的概率分布中采样来合成新颖的树状三维物体。我们在真实和合成植物及植物树木上展示了所提框架的性能和实用性，并表明其显著优于现有的最先进技术。
cs.CV / 31 / 2607.13458
2D Rotary Position Embedding for Scene Text Recognition with Transformers
用于场景文本识别的二维旋转位置嵌入与变换器
Abstract
Scene Text Recognition (STR) remains challenging due to the diversity of text appearances, including curvature, rotation, and perspective distortion. Recent Transformer-based approaches perform well but usually rely on one-dimensional positional encodings that ignore the 2D spatial structure of text images. Axial 2D extensions of Rotary Position Embedding (RoPE) exist for vision Transformers, but they assume roughly square, isotropic image content and apply the rotation only within encoder self-attention. Scene text violates both assumptions: crops are markedly anisotropic, and STR models are encoder-decoder, so the decoder must relate its queries to the encoder's 2D layout through cross-attention. We introduce 2D-RoPE-STR, which adapts axial 2D-RoPE to this setting through (1) an anisotropic row/column dimension allocation matched to the aspect ratio of text, and (2) an extension of the rotary coupling into encoder-decoder cross-attention, letting autoregressive decoding steps attend to encoder tokens by their 2D layout, a setting not addressed by prior encoder-only formulations. Both changes are essentially parameter-free and require no architectural redesign beyond the positional-encoding module. We further introduce a diagnostic protocol (a controlled ablation pair isolating only the positional encoding, an image-level net-win disagreement analysis, and encoder attention visualization) that identifies where and why relative 2D position helps: curved, rotated, and perspective-distorted layouts where reading order departs from a straight horizontal line. On six standard benchmarks (IIIT5K, SVT, ICDAR 2013, ICDAR 2015, CUTE80, SVTP), gains concentrate on exactly these irregular layouts, with ablations isolating each design choice against 1D RoPE and 2D sinusoidal and learnable alternatives.
Chinese Translation
场景文本识别（STR）因文本外观的多样性（包括曲率、旋转和透视失真）而仍然具有挑战性。近期基于变换器的方法表现良好，但通常依赖于一维位置编码，这忽略了文本图像的二维空间结构。虽然存在针对视觉变换器的轴向二维旋转位置嵌入（Rotary Position Embedding, RoPE）扩展，但它们假设图像内容大致为方形且各向同性，并且仅在编码器自注意力中应用旋转。场景文本违反了这两个假设：裁剪的文本明显是各向异性的，且STR模型是编码器-解码器结构，因此解码器必须通过交叉注意力将其查询与编码器的二维布局相关联。我们提出了2D-RoPE-STR，它通过（1）与文本的纵横比相匹配的各向异性行/列维度分配，以及（2）将旋转耦合扩展到编码器-解码器交叉注意力中，从而将轴向二维RoPE适应于这一设置，使自回归解码步骤能够根据编码器的二维布局关注编码器的标记，这一设置在之前的仅编码器公式中未得到解决。这两个变化本质上是无参数的，并且在位置编码模块之外不需要架构重设计。我们进一步引入了一种诊断协议（一个控制消融对，单独隔离位置编码、图像级净收益不一致分析和编码器注意力可视化），以识别相对二维位置有助于的地方和原因：在阅读顺序偏离直水平线的曲线、旋转和透视失真的布局上。在六个标准基准（IIIT5K、SVT、ICDAR 2013、ICDAR 2015、CUTE80、SVTP）上，性能提升恰好集中在这些不规则布局上，消融实验隔离了每个设计选择与一维RoPE及二维正弦和可学习替代方案的对比。
cs.CV / 32 / 2607.13460
LPM: Industrial-Scale Generative Video Restoration
LPM：工业规模的生成视频修复
Abstract
We present the Large Processing Model (LPM), a diffusion-based generative framework for photorealistic video restoration under complex, in-the-wild degradations. To our knowledge, LPM is the first generative video restoration model deployed at industrial scale. LPM addresses the diverse degradations in user-generated content (UGC) through a unified system encompassing large-scale data engineering, foundation-model training, and efficient inference. Its enhanced architecture, progressive training strategy, and temporal-pyramid inference mechanism jointly enable high-fidelity, temporally consistent restoration of arbitrarily long videos across the broad content distribution encountered on UGC platforms. LPM has been deployed in production at Kuaishou, where videos processed by the model account for approximately 45% of total viewing time, delivering consistent improvements across key quality-of-experience metrics. Beyond perceptual enhancement, LPM delivers substantial system-level benefits: at comparable perceptual quality, it reduces bitrate by 20% relative to Kuaishou's in-house codec, yielding annual bandwidth cost savings on the order of hundreds of millions. Its low serving cost also enables integration into products such as Kling, demonstrating that generative restoration can be practical, scalable, and cost-effective for large-scale video processing.
Chinese Translation
我们提出了大型处理模型（LPM），这是一种基于扩散的生成框架，用于在复杂的自然环境退化下进行逼真的视频修复。据我们所知，LPM是首个在工业规模上部署的生成视频修复模型。LPM通过一个统一的系统解决用户生成内容（UGC）中的多样化退化问题，该系统涵盖大规模数据工程、基础模型训练和高效推理。其增强的架构、渐进式训练策略和时间金字塔推理机制共同实现了对UGC平台上广泛内容分布的任意长度视频的高保真、时间一致的修复。LPM已在快手（Kuaishou）投入生产，经过该模型处理的视频占总观看时间的约45%，在关键体验质量指标上提供了一致的改进。除了感知增强，LPM还带来了显著的系统级效益：在可比的感知质量下，相较于快手的内部编解码器，它将比特率降低了20%，每年节省的带宽成本达数亿。其低服务成本还使其能够集成到Kling等产品中，证明生成修复在大规模视频处理中的实用性、可扩展性和成本效益。
cs.CV / 33 / 2607.13468
HIVE-3D: Hierarchical Voxel Enhancement for High-Quality 3D Scene Generation
HIVE-3D：基于层次体素增强的高质量3D场景生成
Abstract
Recently, a line of works can generate impressive 3D objects from a single image, but they are limited by restricted representation resolution, making them unsuitable for 3D scene generation. In this work, we introduce HIVE-3D, a novel method for high-quality 3D scene generation based on hierarchical voxel enhancement framework. Specifically, given a single scene image as input, we first produce a coarse initial scene, then introduce image segmentation and attention-based retrieval to align 2D image components with 3D scene components. Subsequently, we organize these scene relations into a hierarchical component tree, where nodes closer to the leaves denote finer-grained components. Finally, we propose a voxel super-resolution model that generates refined voxels for the target instance while maintaining strong consistency with the coarse voxels. Equipped with this model, we perform coarse-to-fine hierarchical super-resolution on images and voxels for each component, producing a high-resolution and high-quality 3D scene. Extensive experiments demonstrate that our method significantly outperforms previous approaches, achieving state-of-the-art performance.
Chinese Translation
近年来，一系列研究能够从单幅图像生成令人印象深刻的3D物体，但由于受限于表示分辨率，这些方法不适合用于3D场景生成。在本研究中，我们提出了HIVE-3D，一种基于层次体素增强框架的高质量3D场景生成新方法。具体而言，给定一幅场景图像作为输入，我们首先生成一个粗略的初始场景，然后引入图像分割和基于注意力的检索，将2D图像组件与3D场景组件对齐。随后，我们将这些场景关系组织成一个层次组件树，其中靠近叶子的节点表示更细粒度的组件。最后，我们提出了一种体素超分辨率模型，该模型在保持与粗体素强一致性的同时，为目标实例生成精细体素。借助该模型，我们对每个组件的图像和体素执行粗到细的层次超分辨率，生成高分辨率和高质量的3D场景。大量实验表明，我们的方法显著优于之前的研究，达到了最先进的性能。
cs.CV / 34 / 2607.13471
Bring Music The Horizon: Music-Driven 360$^\circ$ Video Generation
带来音乐的视野：基于音乐的360$^ ext{°}$视频生成
Abstract
Music visualization offers a powerful way to enhance listeners' understanding and experience of music by translating auditory signals into visual forms. However, most existing approaches either rely heavily on lyrics or generate flat, non-immersive videos similar to conventional music videos, which limits their ability to convey the emotional dynamics of music and provide an immersive listening experience. We propose Bring Music The Horizon, an emotion-aware pipeline for music-driven 360$^\circ$ video generation. Given an input song, our work first estimates its emotional trajectory by predicting valence-arousal values at the level of every four bars. These values are then converted into emotion-aware visual guidance using EmotiCrafter, and these guidance vectors can be manipulated by the SEGA framework, which provides fine-grained semantic control for keyframe generation. Finally, image-to-video models are applied to the generated keyframes to synthesize temporally continuous 360$^\circ$ videos for immersive music visualization. Our pipeline generates 360$^\circ$ music visualization videos that reflect the emotional progression and temporal structure of the input song. We demonstrate its capability using songs from different genres and provide qualitative comparisons with From-Sound-To-Sight, a representative audio-to-visual generation baseline, on our project page at https://etoile-et-toi-mp3.github.io/BMTH_Project_Page/.
Chinese Translation
音乐可视化为增强听众对音乐的理解和体验提供了一种强有力的方式，通过将听觉信号转化为视觉形式。然而，大多数现有方法要么过于依赖歌词，要么生成类似于传统音乐视频的平面、非沉浸式视频，这限制了它们传达音乐情感动态和提供沉浸式听觉体验的能力。我们提出了“带来音乐的视野”（Bring Music The Horizon），这是一个情感感知的音乐驱动360$^ ext{°}$视频生成管道。给定一首输入歌曲，我们的工作首先通过预测每四小节的愉悦度-唤醒度（valence-arousal）值来估计其情感轨迹。这些值随后通过EmotiCrafter转化为情感感知的视觉指导，这些指导向量可以通过SEGA框架进行操作，从而为关键帧生成提供细粒度的语义控制。最后，将图像到视频模型应用于生成的关键帧，以合成时间连续的360$^ ext{°}$视频，实现沉浸式音乐可视化。我们的管道生成的360$^ ext{°}$音乐可视化视频反映了输入歌曲的情感进程和时间结构。我们使用不同类型的歌曲展示了其能力，并在我们的项目页面https://etoile-et-toi-mp3.github.io/BMTH_Project_Page/上与代表性的音频到视觉生成基线“从声音到视觉”（From-Sound-To-Sight）进行了定性比较。
cs.CV / 35 / 2607.13481
GPOcc++: Unified Sparse Gaussian Occupancy Prediction with Visual Geometry Priors
GPOcc++：结合视觉几何先验的统一稀疏高斯占用预测
Abstract
Accurate 3D scene understanding is fundamental to embodied intelligence and autonomous driving, where 3D occupancy provides a unified representation of objects, structures, and free space. However, recovering such a complete volumetric representation from visual observations remains challenging, particularly in occluded and unobserved regions. Visual geometry priors offer strong and generalizable geometric cues for addressing this challenge, but their outputs are inherently surface-centric, whereas occupancy prediction requires reasoning about volumetric interiors and free space. To bridge this gap, we introduce GPOcc, which transforms visual geometry priors into occupancy-aware sparse Gaussian representations for efficient and expressive volumetric scene modeling. Building on GPOcc, GPOcc++ models multi-view observations and temporal sequences within a unified framework, allowing spatial and temporal evidence to be handled through the same representation. We further extend GPOcc++ from indoor scenes to outdoor occupancy prediction. Extensive experiments on both indoor and outdoor benchmarks demonstrate consistently strong performance across both multi-view and temporal settings, together with favorable efficiency and generalization. Code will be released at https://github.com/JuIvyy/GPOcc.
Chinese Translation
准确的三维场景理解是具身智能和自动驾驶的基础，其中三维占用提供了对象、结构和自由空间的统一表示。然而，从视觉观测中恢复这样一个完整的体积表示仍然具有挑战性，特别是在被遮挡和未观察到的区域。视觉几何先验为解决这一挑战提供了强大且可推广的几何线索，但它们的输出本质上是以表面为中心的，而占用预测需要对体积内部和自由空间进行推理。为了解决这一问题，我们引入了GPOcc，它将视觉几何先验转化为占用感知的稀疏高斯表示，以实现高效且富有表现力的体积场景建模。在GPOcc的基础上，GPOcc++在统一框架内建模多视角观测和时间序列，允许通过相同的表示处理空间和时间证据。我们进一步将GPOcc++从室内场景扩展到室外占用预测。在室内和室外基准上的大量实验表明，在多视角和时间序列设置中均表现出一致的强大性能，同时具有良好的效率和泛化能力。代码将发布在 https://github.com/JuIvyy/GPOcc。
cs.CV / 36 / 2607.13492
CASA-SDF: Curriculum-Aware Spatial Adaptation with Curvature-Guided Density for Neural Implicit Surface Reconstruction
CASA-SDF：基于课程意识的空间适应与曲率引导密度用于神经隐式表面重建
Abstract
Neural implicit representations have emerged as a powerful paradigm for 3D reconstruction. However, high-fidelity indoor surface reconstruction remains a significant challenge, primarily due to the pronounced \emph{geometric heterogeneity} of indoor scenes. Large texture-less planar regions typically require stronger regularization to suppress high-frequency artifacts, while thin structures demand sharper, more adaptive representations to mitigate the spectral bias of multi-layer perceptrons (MLPs) and prevent over-smoothing. Existing approaches often rely on spatially indiscriminate prior supervision and a scene-global SDF-to-density transformation, which constrains their ability to balance planar smoothness and detail preservation. In this paper, we propose CASA-SDF (Curriculum-Aware Spatial Adaptation for SDF), a unified framework that addresses this challenge via complementary adaptations of supervision and representation capacity. Specifically, Hybrid Spatially-Adaptive Uncertainty Annealing (SAUA) fuses semantic and photometric uncertainties to construct a pixel-wise curriculum for monocular prior supervision. This strategy maintains regularization in reliable regions while attenuating unreliable supervision early in training to enable data-driven photometric refinement. Meanwhile, Curvature-Aware Locally Adaptive Density Transformation (CALADT) progressively modulates the sharpness of the SDF-to-density mapping via a curvature proxy to enhance the representation of thin structures. Extensive experiments on benchmark indoor datasets demonstrate that CASA-SDF improves surface completeness and detail recovery on high-frequency structures, without compromising the stability of planar surfaces.
Chinese Translation
神经隐式表示已成为三维重建的强大范式。然而，高保真室内表面重建仍然是一个重大挑战，主要由于室内场景的显著 extit{几何异质性}。大面积无纹理的平面区域通常需要更强的正则化来抑制高频伪影，而细结构则需要更尖锐、更具适应性的表示，以减轻多层感知器（MLPs）的谱偏差并防止过度平滑。现有方法通常依赖于空间上不加区分的先验监督和全局场景的SDF到密度转换，这限制了它们在平面光滑性与细节保留之间的平衡能力。本文提出了CASA-SDF（基于课程意识的SDF空间适应），这是一个统一框架，通过监督和表示能力的互补适应来解决这一挑战。具体而言，混合空间自适应不确定性退火（SAUA）融合了语义和光度不确定性，以构建单目先验监督的逐像素课程。这一策略在可靠区域保持正则化，同时在训练初期减弱不可靠的监督，以实现数据驱动的光度细化。同时，曲率感知局部自适应密度转换（CALADT）通过曲率代理逐步调节SDF到密度映射的锐度，以增强细结构的表示。在基准室内数据集上的大量实验表明，CASA-SDF在高频结构的表面完整性和细节恢复方面有所改善，同时不影响平面表面的稳定性。
cs.CV / 37 / 2607.13499
M2P-AD: Memory-to-Prototype Learning with Boundary-aware Score Refinement for 3D Anomaly Detection
M2P-AD：基于边界感知评分细化的记忆到原型学习用于 3D 异常检测
Abstract
3D anomaly detection has recently emerged as an important research topic in computer vision. Although existing methods have achieved high performance, excessive anomaly responses in normal regions and false positives near object boundaries remain unresolved challenges. To address these challenges, we propose a novel 3D anomaly detection model, Memory-to-Prototype Anomaly Detection (M2P-AD), which effectively models the distribution of normal features while suppressing excessive anomaly scores in normal regions and false positives near object boundaries. Specifically, we introduce a Memory-to-Prototype (M2P) module that learns representative prototypes from normal feature embeddings to preserve important structural information of objects. In addition, a Boundary extraction (BE) module is integrated to identify object boundaries, and a Boundary-aware score refinement (BSR) strategy is applied to recalibrate anomaly scores by incorporating boundary characteristics. The proposed method is evaluated on Real3D-AD, Anomaly-ShapeNet, and MulSen-AD, achieving state-of-the-art performance. Qualitative results demonstrate that excessive anomaly scores in normal regions are reduced and false positives near object boundaries are suppressed, resulting in more accurate and stable anomaly localization. The results indicate that the proposed approach enables more reliable 3D anomaly detection and provides a robust solution applicable to real-world industrial environments.
Chinese Translation
3D 异常检测最近已成为计算机视觉领域的重要研究课题。尽管现有方法已取得高性能，但在正常区域的过度异常响应和物体边界附近的假阳性仍然是未解决的挑战。为了解决这些问题，我们提出了一种新颖的 3D 异常检测模型——记忆到原型异常检测（Memory-to-Prototype Anomaly Detection, M2P-AD），该模型有效地建模正常特征的分布，同时抑制正常区域的过度异常评分和物体边界附近的假阳性。具体而言，我们引入了一个记忆到原型（Memory-to-Prototype, M2P）模块，该模块从正常特征嵌入中学习代表性原型，以保留物体的重要结构信息。此外，集成了一个边界提取（Boundary extraction, BE）模块来识别物体边界，并应用边界感知评分细化（Boundary-aware score refinement, BSR）策略，通过结合边界特征来重新校准异常评分。我们在 Real3D-AD、Anomaly-ShapeNet 和 MulSen-AD 数据集上评估了所提出的方法，取得了最先进的性能。定性结果表明，正常区域的过度异常评分得到了降低，物体边界附近的假阳性得到了抑制，从而实现了更准确和稳定的异常定位。结果表明，所提出的方法使得 3D 异常检测更加可靠，并为适用于现实工业环境提供了稳健的解决方案。
cs.CV / 38 / 2607.13500
Attention-Free and Lightweight Token Reduction for Efficient Vision-Language Models
无注意力且轻量级的令牌减少方法以提高视觉-语言模型的效率
Abstract
Vision-Language Models (VLMs) have achieved strong performance in multimodal understanding, yet remain challenging to deploy on resource-constrained edge devices due to the substantial computational overhead of processing numerous visual tokens. Token reduction is a promising direction for accelerating VLMs inference, but existing approaches either rely on attention maps that are incompatible with modern acceleration frameworks or depend on computationally intensive pairwise similarity comparisons, which undermine scalability and negate their practical benefits in deployment. In this paper, we propose an attention-free and lightweight token reduction framework as a plug-and-play module for VLMs, which preserves both important and diverse tokens to produce a compact visual representation. First, to enable attention-free importance estimation, we adopt an information-theoretic perspective and quantify token information using a novel entropy-based criterion, retaining those with more expressive and less degenerate feature representations. Second, to ensure diverse visual coverage in a lightweight manner, we introduce a transformation-induced consistency signal where similar tokens yield similar signals, such that sorting by this signal places similar tokens close to each other and enables stride-based selection to produce a diverse token set. Extensive experiments across multiple VLMs benchmarks demonstrate that our framework achieves a favorable accuracy-efficiency trade-off, maintaining competitive performance under aggressive compression.
Chinese Translation
视觉-语言模型（VLMs）在多模态理解方面取得了强劲的性能，但由于处理大量视觉令牌所需的显著计算开销，仍然难以在资源受限的边缘设备上部署。令牌减少是加速VLM推理的一个有前景的方向，但现有方法要么依赖于与现代加速框架不兼容的注意力图，要么依赖于计算密集的成对相似性比较，这削弱了可扩展性并否定了它们在部署中的实际效益。本文提出了一种无注意力且轻量级的令牌减少框架，作为VLM的即插即用模块，能够保留重要且多样的令牌，从而生成紧凑的视觉表示。首先，为了实现无注意力的重要性估计，我们采用信息论的视角，通过一种新颖的基于熵的标准量化令牌信息，保留那些具有更具表现力和较少退化特征表示的令牌。其次，为了以轻量级的方式确保多样的视觉覆盖，我们引入了一种由变换引起的一致性信号，其中相似的令牌产生相似的信号，从而通过该信号排序使相似的令牌彼此靠近，并实现基于步幅的选择以生成多样的令牌集。在多个VLM基准上的广泛实验表明，我们的框架实现了良好的准确性与效率的权衡，在激进压缩下保持了竞争力的性能。
cs.CV / 39 / 2607.13504
DP-BOA: Dirichlet-Process Birth-or-Assign for On-the-Fly Category Discovery
DP-BOA：用于即时类别发现的狄利克雷过程出生或分配方法
Abstract
On-the-fly category discovery requires deciding for each incoming test sample whether to assign it to an existing category or spawn a new one. Existing methods typically implement this decision through matching-based heuristics, such as radius- or hash-based rules. While effective in practice, these methods usually treat category birth implicitly as a fallback when no existing category matches confidently, rather than as an explicit alternative supported by its own statistical evidence. To address this, we propose DP-BOA, a posterior-predictive decision framework based on an online Dirichlet-process Gaussian mixture model with a Normal-Inverse-Wishart prior. During training, we use labeled data to calibrate a shared NIW prior over category Gaussians and warm-start the known-category posteriors. At test time, for each incoming sample, DP-BOA compares the posterior predictive evidence for assignment to existing categories against the evidence for spawning a new category induced by the DP prior, and then updates category statistics online after the decision. The method captures anisotropic category geometry and naturally adapts decision confidence as evidence accumulates. Across standard OCD benchmarks, DP-BOA consistently outperforms strong baselines and delivers particularly strong novel-class discovery performance while maintaining competitive known-class accuracy.
Chinese Translation
即时类别发现需要为每个到来的测试样本决定是将其分配到现有类别还是生成一个新类别。现有的方法通常通过基于匹配的启发式规则来实现这一决策，例如基于半径或哈希的规则。尽管在实践中有效，这些方法通常将类别的生成隐式视为一种后备选项，当没有现有类别能够自信地匹配时，而不是作为一种由自身统计证据支持的明确替代方案。为了解决这个问题，我们提出了DP-BOA，一种基于在线狄利克雷过程高斯混合模型（Dirichlet-process Gaussian mixture model）和正态-逆威沙特先验（Normal-Inverse-Wishart prior）的后验预测决策框架。在训练过程中，我们使用标记数据来校准类别高斯的共享NIW先验，并为已知类别的后验分布提供热启动。在测试时，对于每个到来的样本，DP-BOA比较分配到现有类别的后验预测证据与由DP先验诱导的新类别生成证据，然后在决策后在线更新类别统计信息。该方法捕捉了各向异性的类别几何形状，并随着证据的积累自然调整决策信心。在标准的在线类别发现（OCD）基准测试中，DP-BOA始终优于强基线，并在保持竞争的已知类别准确性的同时，提供了特别强的新增类别发现性能。
cs.CV / 40 / 2607.13506
TRACE-PCa: Predicting Prostate Cancer Progression from Longitudinal MRI During Active Surveillance
TRACE-PCa：基于纵向MRI预测前列腺癌在主动监测期间的进展
Abstract
Active surveillance (AS) is the preferred strategy for favorable-risk prostate cancer, yet current protocols rely on scheduled repeat biopsies, most of which reveal no progression and are unnecessary. Existing risk-stratification tools operate on single time-point imaging or depend on explicit lesion segmentation, limiting their ability to capture longitudinal change and excluding patients without an MRI-visible lesion. In this study, we propose an end-to-end temporal and multimodal model for predicting pathological progression during AS without lesion segmentation. We encode each serial scan with a pretrained 3D MRI foundation model and introduce a temporal attention gate that recalibrates the multi-visit features to amplify focal imaging changes associated with progression. The gated imaging representation is then fused with clinical variables in a multimodal framework to estimate the probability of progression. Validated on a longitudinal AS cohort, our approach consistently outperforms competing baselines and performs comparably to the radiologist assessment representing current clinical practice. It maintains high negative predictive value while achieving higher positive predictive value, demonstrating its potential to safely reduce unnecessary biopsies during surveillance.
Chinese Translation
主动监测（AS）是针对良性风险前列腺癌的首选策略，但目前的协议依赖于定期重复活检，而大多数活检结果并未显示病情进展，且这些活检是多余的。现有的风险分层工具基于单一时间点的影像或依赖于明确的病灶分割，限制了其捕捉纵向变化的能力，并排除了没有MRI可见病灶的患者。在本研究中，我们提出了一种端到端的时间序列和多模态模型，用于在不进行病灶分割的情况下预测AS期间的病理进展。我们使用预训练的3D MRI基础模型对每个序列扫描进行编码，并引入一个时间注意力门，重新校准多次访问特征，以增强与进展相关的局部影像变化。然后，将门控影像表示与临床变量融合在一个多模态框架中，以估计进展的概率。在一个纵向AS队列中进行验证，我们的方法始终优于竞争基线，并且与代表当前临床实践的放射科医师评估表现相当。它保持了高的阴性预测值，同时实现了更高的阳性预测值，展示了其在监测期间安全减少不必要活检的潜力。
cs.CV / 41 / 2607.13515
DriveFace: A Cross-Spectral Through-Glass Face Dataset for On-the-Move Vehicular Border Control
DriveFace：一种用于移动车辆边境控制的跨光谱透玻璃人脸数据集
Abstract
The continuous growth in cross-border mobility places increasing pressure on existing border control infrastructures, motivating on-the-move biometric authentication, in which travellers are identified directly inside their vehicles at checkpoints. Face recognition is well-suited to this setting, as it can be acquired passively and at a distance. Its development, however, is hindered by the lack of representative datasets: existing benchmarks are collected in controlled environments and do not capture the challenges inherent to vehicular acquisition, including motion blur, variable illumination, occlusions, and cross-spectral enrollment. To address this gap, we introduce a dataset for on-the-move face recognition in border-control scenarios, comprising NIR vehicle-crossing videos paired with smartphone-based pre-enrollment data. Baseline evaluations with state-of-the-art models show clear performance limitations under these realistic conditions, highlighting the need for dedicated methods to advance the field.
Chinese Translation
跨境流动的持续增长对现有边境控制基础设施施加了越来越大的压力，这促使了移动生物识别认证的发展，在这种情况下，旅客在检查站内直接在其车辆内被识别。人脸识别非常适合这种场景，因为它可以被被动地和远距离地获取。然而，其发展受到代表性数据集缺乏的制约：现有基准是在受控环境中收集的，未能捕捉到车辆采集固有的挑战，包括运动模糊、光照变化、遮挡和跨光谱注册。为了解决这一空白，我们引入了一种用于边境控制场景中移动人脸识别的数据集，包含与智能手机预注册数据配对的近红外（NIR）车辆穿越视频。与最先进模型的基线评估显示，在这些现实条件下存在明显的性能限制，突显了推动该领域发展的专门方法的必要性。
cs.CV / 42 / 2607.13527
VGIF-Score: Interpretable and Diagnostic Evaluation of Spatio-Temporal Instruction Following in Video Generation
VGIF-Score：可解释的时空指令遵循在视频生成中的诊断评估
Abstract
Recent video generation models (VGMs) have made substantial progress in visual fidelity, yet their ability to follow long, compositional instructions remains insufficiently evaluated. Existing evaluation protocols often rely on prompts that are short and semantically shallow, with limited atomic constraints and weak spatio-temporal dependencies. They also frequently depend on costly human evaluation or handcrafted vision pipelines, while providing little diagnostic insight into which instruction constraints succeed or fail. To address this gap, we propose VGIF-Score, a highly automated and interpretable framework for evaluating instruction following in video generation. VGIF-Score consists of two complementary components: an objective completion branch that parses prompts into a Spatio-Temporal Directed Acyclic Graph (ST-DAG) and performs dependency-aware QA with short-circuit diagnostics, and a subjective satisfaction branch that uses instruction-conditioned AutoRubric to assess cinematography, visual purity, motion smoothness, and physics adherence. Together, these components produce a unified score that captures both objective completion and perceptual satisfaction. We instantiate this framework on VGIF-Bench, a benchmark of 223 long, structurally entangled prompts paired with approximately 4.3K fine-grained evaluation items. Experiments on 14 proprietary and open-source VGMs across more than 3K generated videos show that VGIF-Score provides reliable, interpretable, and diagnostically useful evaluation of video generation instruction following. The code will be available at https://github.com/PRIS-CV/VGIF-SCORE.
Chinese Translation
近期的视频生成模型（VGM）在视觉逼真度方面取得了显著进展，但它们遵循长篇、复杂指令的能力仍然缺乏充分评估。现有的评估协议通常依赖于短小且语义浅显的提示，具有有限的原子约束和较弱的时空依赖性。同时，它们往往依赖于昂贵的人力评估或手工制作的视觉管道，且对哪些指令约束成功或失败提供的诊断见解有限。为了解决这一问题，我们提出了VGIF-Score，一个高度自动化且可解释的框架，用于评估视频生成中的指令遵循。VGIF-Score由两个互补的组成部分构成：一个客观完成分支，将提示解析为时空有向无环图（ST-DAG），并通过短路诊断执行依赖感知的质量评估；一个主观满意度分支，使用基于指令的AutoRubric来评估摄影、视觉纯度、运动平滑度和物理遵循性。这两个组成部分共同生成一个统一的评分，捕捉客观完成和感知满意度。我们在VGIF-Bench上实现了这一框架，这是一个包含223个长篇、结构复杂的提示及约4.3K个细粒度评估项目的基准。对14个专有和开源VGM在超过3K个生成视频上的实验表明，VGIF-Score提供了可靠、可解释且具有诊断价值的视频生成指令遵循评估。代码将发布于 https://github.com/PRIS-CV/VGIF-SCORE。
cs.CV / 43 / 2607.13539
ThinkBLOX: 3D Indoor Scene Generation with Progressive Reasoning
ThinkBLOX：基于渐进推理的3D室内场景生成
Abstract
While traditional graphics methods often synthesize 3D indoor scenes autoregressively or hierarchically, recent vision-language model (VLM)-based generators predominantly adopt a one-shot paradigm where the full layout is planned at once. This one-shot approach often requires global re-optimization or complete reconstruction during interactive editing (e.g., inserting or moving objects) and can lead to physically or semantically poorly organized arrangements. To address these challenges, we propose ThinkBLOX, a VLM-based progressive reasoning framework that iteratively designs and refines 3D scenes. ThinkBLOX treats layout generation as a state-conditioned, step-by-step reasoningand-action process. To power this, we construct the ThinkBLOX-Data-200K dataset, containing 224,757 procedural placement pairs annotated with multi-view scene context, explicit Chain-of-Thought (CoT) rationales, and structured JSON layouts. Through supervised fine-tuning (SFT) on this dataset, the VLM learns to bridge the reasoning-action gap under incremental updates. Furthermore, recognizing that scene synthesis is inherently a multisolution task where SFT suffers from reward conflict, we introduce Tier-Decoupled GDPO. This reinforcement learning scheme organizes heterogeneous rewards into distinct tiers, stabilizing policy optimization across physical validity, semantic plausibility, and reasoning-action consistency. Extensive experiments show that ThinkBLOX significantly outperforms recent one-shot and iterative baselines in physical plausibility, semantic alignment, and interactive editability. Additionally, we show that it supports diverse applications, including both global and local generation and rearrangement of 3D scenes.
Chinese Translation
传统图形方法通常以自回归或分层的方式合成3D室内场景，而最近基于视觉-语言模型（VLM）的生成器主要采用一次性范式，在此范式中，整个布局一次性规划完成。这种一次性方法在交互编辑（例如，插入或移动物体）过程中往往需要进行全局重新优化或完全重建，可能导致物理或语义上组织不良的排列。为了解决这些挑战，我们提出了ThinkBLOX，一种基于VLM的渐进推理框架，能够迭代设计和优化3D场景。ThinkBLOX将布局生成视为一个状态条件下的逐步推理与行动过程。为此，我们构建了ThinkBLOX-Data-200K数据集，其中包含224,757对程序化放置样本，并附有多视角场景上下文、明确的思维链（Chain-of-Thought, CoT）推理和结构化的JSON布局。通过对该数据集的监督微调（SFT），VLM学习在增量更新下弥合推理与行动之间的差距。此外，考虑到场景合成本质上是一个多解任务，而SFT在此过程中面临奖励冲突，我们引入了层解耦的GDPO（Tier-Decoupled GDPO）。该强化学习方案将异质奖励组织成不同层级，从而在物理有效性、语义合理性和推理-行动一致性之间稳定策略优化。大量实验表明，ThinkBLOX在物理合理性、语义对齐和交互可编辑性方面显著优于最近的一次性和迭代基线。此外，我们还展示了它支持多种应用，包括3D场景的全局和局部生成及重新排列。
cs.CV / 44 / 2607.13563
Nexus: Native Mesh Generation with Diffusion
Nexus：基于扩散的原生网格生成
Abstract
Generating high-quality triangle meshes is essential for film, gaming, and interactive 3D applications. Mainstream methods rely on mesh serialization and autoregressive processes, which stuggles in effective inference and is sensitive to error accumulation. In this paper, we present Nexus, a diffusion method that achieves holistic mesh generation via decoupled vertex and topology generation. First, we view mesh vertices as sparse voxels organized as an octree and adopt a diffusion model to generate the vertices in a coarse-to-fine manner. Second, for topology modeling, we propose Spacetime Interval, as an extension of Spacetime Distance to encode arbitrary edge and face topology into continuous per-vertex embeddings. It allows for a global and efficient recovery of complex topology. We then employ a diffusion model to generate the continuous embeddings on the generated vertices. Extensive experiments on the Objaverse and Toys4K datasets and in-the-wild images demonstrate that our method outperforms state-of-the-art autoregressive and two-stage baselines, effectively circumventing the inherent limitations of sequential mesh modeling. A blind user study from 3D practitioners confirms strong perceptual preference for our results.
Chinese Translation
生成高质量的三角网格对于电影、游戏和交互式3D应用至关重要。主流方法依赖于网格序列化和自回归过程，这在有效推断方面存在困难，并且对误差积累敏感。本文提出了Nexus，一种通过解耦的顶点和拓扑生成实现整体网格生成的扩散方法。首先，我们将网格顶点视为组织成八叉树的稀疏体素，并采用扩散模型以粗到细的方式生成顶点。其次，对于拓扑建模，我们提出了时空区间（Spacetime Interval），作为时空距离（Spacetime Distance）的扩展，将任意边和面拓扑编码为连续的每顶点嵌入。这允许对复杂拓扑进行全局和高效的恢复。然后，我们在生成的顶点上应用扩散模型生成连续嵌入。在Objaverse和Toys4K数据集以及现实世界图像上的大量实验表明，我们的方法在性能上优于最先进的自回归和两阶段基线，有效规避了顺序网格建模的固有限制。一项来自3D从业者的盲测用户研究确认了对我们结果的强烈感知偏好。
cs.CV / 45 / 2607.13569
GHR-VLM: Making Zero-Shot Transit Video Analytics Realizable with Grounded Hybrid Reasoning
GHR-VLM：通过基于视觉的混合推理实现零-shot公交视频分析
Abstract
Transit video understanding can provide valuable fine-grained data that conventional passenger counters and fare systems cannot capture. However, supervised video models require task-specific annotations, while applying vision-language models (VLMs) directly to long onboard videos is unreliable and costly. To leverage the complementary strengths of both approaches, we propose GHR-VLM, a visual grounded hybrid reasoning framework for zero-shot transit-bus video analytics. It is motivated by the observation that explicit visual grounding can improve VLM reasoning by converting long surveillance streams into compact, passenger-centered spatiotemporal evidence. Specifically, we propose an edge-cloud design in which a lightweight edge-based monitor continuously tracks door status and segments passenger clips. A backend VLM then identifies boarding passengers and classifies payment behavior through a two-stage coarse-to-fine refinement of spatiotemporal evidence. By invoking the VLM only on grounded passenger clips and contact sheets, GHR-VLM reduces cloud inference, avoids payment-specific training data, and supplies the localized evidence that VLMs otherwise struggle to identify. Evaluation on 486 minutes of real-world bus surveillance video demonstrates the potential of grounded edge-cloud reasoning for passenger-level payment analytics while highlighting the challenges posed by degraded video conditions.
Chinese Translation
公交视频理解可以提供传统乘客计数器和票务系统无法捕捉的有价值的细粒度数据。然而，监督视频模型需要特定任务的标注，而直接将视觉语言模型（VLMs）应用于长时间的车载视频则不可靠且成本高昂。为了利用这两种方法的互补优势，我们提出了GHR-VLM，一种用于零-shot公交视频分析的视觉基础混合推理框架。其动机在于观察到显式的视觉基础可以通过将长时间监控流转换为紧凑的、以乘客为中心的时空证据来改善VLM推理。具体而言，我们提出了一种边缘-云设计，其中一个轻量级的边缘监控器持续跟踪门状态并分割乘客视频片段。随后，后端VLM通过对时空证据进行两阶段的粗到细的精炼，识别上车乘客并分类支付行为。通过仅在基础乘客片段和接触表上调用VLM，GHR-VLM减少了云推理，避免了特定支付的训练数据，并提供了VLM通常难以识别的局部证据。在486分钟的真实公交监控视频上的评估展示了基于视觉的边缘-云推理在乘客级支付分析中的潜力，同时突显了退化视频条件带来的挑战。
cs.CV / 46 / 2607.13586
UniPhysGen: Unified Physical Grounding for Simulation-Ready 3D Assets
UniPhysGen：用于模拟准备的3D资产的统一物理基础
Abstract
Physically grounded 3D assets are increasingly important for embodied AI and robotic simulation. However, most existing 3D assets lack unified physical semantics, including articulation semantics and intrinsic physical properties, required for realistic interaction. Current approaches either treat these semantics independently or rely on canonicalized object structures, limiting robustness across heterogeneous 3D assets. We present UniPhys, a scalable framework for automatically transforming raw 3D assets into simulation-ready assets with unified physical semantics. Based on UniPhys, we construct UniPhys-40K, a large-scale physically grounded dataset, together with UniPhys-Bench, a carefully verified benchmark for unified physical grounding evaluation. We further introduce UniPhysGen, a unified physical grounding model that jointly reasons over articulation semantics and intrinsic physical properties. UniPhysGen incorporates geometry-robust articulation grounding to mitigate geometric shortcut bias under heterogeneous part decompositions. Extensive experiments demonstrate state-of-the-art performance across articulation grounding and intrinsic physical property estimation tasks, while the resulting assets can be directly deployed in robotic simulation environments for realistic physical interaction. Our code and dataset will be available at https://github.com/breezexian/UniPhysGen.
Chinese Translation
物理基础的3D资产在具身人工智能和机器人仿真中变得越来越重要。然而，大多数现有的3D资产缺乏统一的物理语义，包括实现真实交互所需的关节语义和内在物理属性。目前的方法要么独立处理这些语义，要么依赖于规范化的对象结构，这限制了在异构3D资产中的鲁棒性。我们提出了UniPhys，这是一个可扩展的框架，用于自动将原始3D资产转换为具有统一物理语义的模拟准备资产。基于UniPhys，我们构建了UniPhys-40K，这是一个大规模的物理基础数据集，以及UniPhys-Bench，这是一个经过仔细验证的统一物理基础评估基准。我们进一步介绍了UniPhysGen，这是一个统一的物理基础模型，能够共同推理关节语义和内在物理属性。UniPhysGen结合了几何鲁棒的关节基础，以减轻在异构部件分解下的几何捷径偏差。大量实验表明，在关节基础和内在物理属性估计任务中，UniPhysGen表现出最先进的性能，同时生成的资产可以直接在机器人仿真环境中部署，以实现真实的物理交互。我们的代码和数据集将可在 https://github.com/breezexian/UniPhysGen 获取。
cs.CV / 47 / 2607.13639
OvisOCR2 Technical Report
OvisOCR2 技术报告
Abstract
We introduce OvisOCR2, a 0.8B document parsing model. OvisOCR2 is designed as an end-to-end parser: given a document page image, it generates a Markdown representation in natural reading order, covering text, formulas, tables, and visual regions. We build a data engine that combines filtered real-document annotations with synthetic pages whose rendered images and Markdown targets are derived from the same HTML source. The training recipe includes supervised fine-tuning, reinforcement learning on a 4B branch with a multi-component reward design, on-policy distillation into the 0.8B model, and model fusion. On OmniDocBench v1.6, OvisOCR2 achieves a state-of-the-art overall score of 96.58, placing an end-to-end model at the top of this leaderboard previously dominated by pipeline methods and highlighting the potential of end-to-end document parsing. On PureDocBench, OvisOCR2 also achieves the highest Avg3 score of 75.06. Beyond these two public benchmarks, we evaluate OvisOCR2 on an in-house benchmark designed to cover a broader set of long-tail and challenging scenarios. OvisOCR2 obtains the best overall performance among the compared methods, providing further evidence of its generalization and robustness. OvisOCR2 is available at https://huggingface.co/ATH-MaaS/OvisOCR2.
Chinese Translation
我们介绍了 OvisOCR2，这是一个 0.8B 的文档解析模型。OvisOCR2 被设计为一个端到端的解析器：给定一个文档页面图像，它生成自然阅读顺序的 Markdown 表示，涵盖文本、公式、表格和视觉区域。我们构建了一个数据引擎，将过滤后的真实文档注释与合成页面结合在一起，其渲染图像和 Markdown 目标均源自相同的 HTML。训练方案包括监督微调、在一个 4B 分支上进行的强化学习，采用多组件奖励设计，以及对 0.8B 模型的在线蒸馏和模型融合。在 OmniDocBench v1.6 上，OvisOCR2 达到了 96.58 的最先进的整体得分，将一个端到端模型置于这个之前由管道方法主导的排行榜的顶端，突显了端到端文档解析的潜力。在 PureDocBench 上，OvisOCR2 也达到了最高的 Avg3 得分 75.06。除了这两个公共基准外，我们还在一个内部基准上评估了 OvisOCR2，该基准旨在覆盖更广泛的长尾和具有挑战性的场景。OvisOCR2 在比较的方法中获得了最佳的整体性能，进一步证明了其泛化能力和鲁棒性。OvisOCR2 可在 https://huggingface.co/ATH-MaaS/OvisOCR2 获取。
cs.CV / 48 / 2607.13646
Human4K: A Large-Scale 4K Multi-View Mocap Dataset for Whole-Body 3D Human Reconstruction
Human4K：一个大规模4K多视角动作捕捉数据集用于全身3D人类重建
Abstract
Recent advances in 3D human reconstruction have improved overall performance, yet current models still fail in the most challenging real-world scenarios. They often produce unstable geometry, inaccurate limb articulation and unreliable predictions under depth ambiguity or self-occlusion. A key reason is that existing datasets still lack the combination of high-resolution images, high-precision annotations and diverse whole-body motions required to support robust reconstruction. To address this gap, we present Human4K, a large-scale 4K multi-view whole-body human reconstruction dataset with mocap-accurate SMPL-X annotations. Human4K contains over six million 4K images captured by an eight-view high-resolution camera system synchronized with a professional Vicon motion capture setup, covering 11 subjects performing complex, highly articulated and strongly self-occluded full-body motions. All sequences are processed by a Motion-Retargeting and Refinement Module (MRRM) to ensure precise alignment for the full body and extremities. Experimental results show that training with Human4K consistently improves whole-body reconstruction on standard benchmarks, with particularly large gains for hands, feet and depth-ambiguous limb configurations.
Chinese Translation
近年来，3D人类重建的进展提升了整体性能，但当前模型在最具挑战性的现实场景中仍然表现不佳。它们常常产生不稳定的几何形状、不准确的肢体关节运动，以及在深度模糊或自遮挡情况下的不可靠预测。一个关键原因是现有数据集仍然缺乏高分辨率图像、高精度标注和多样化全身动作的结合，这些都是支持稳健重建所必需的。为了解决这一问题，我们提出了Human4K，一个大规模的4K多视角全身人类重建数据集，具有动作捕捉精度的SMPL-X标注。Human4K包含超过六百万张由八视角高分辨率相机系统捕获的4K图像，这些相机系统与专业的Vicon动作捕捉设备同步，涵盖11个受试者进行复杂、高度关节化和强自遮挡的全身动作。所有序列都经过运动重定向与精炼模块（Motion-Retargeting and Refinement Module, MRRM）处理，以确保全身及四肢的精确对齐。实验结果表明，使用Human4K进行训练在标准基准测试上持续改善全身重建，特别是在手、脚和深度模糊肢体配置方面取得了显著提升。
cs.CV / 49 / 2607.13647
Beyond Color Geometry: Evaluating Human-Like Color Representations in Vision Models
超越颜色几何：评估视觉模型中的类人颜色表示
Abstract
Do vision models see colors the way humans do? Existing evaluations of color representations usually compare them with geometric spaces such as CIELAB or with discrete color labels. These references capture perceptual distance or category membership, but not the graded way in which people organize colors. We evaluate color grounding against a fuzzy perceptual model with 86 graded categories fitted to human survey data. The framework can be applied to any image encoder and measures three complementary properties: category boundaries, category compactness, and graded alignment beyond what color geometry alone can explain. Across eleven Vision Transformer encoders, the category-level results are broadly similar, whereas graded alignment differs substantially. Masked Autoencoders achieve the strongest beyond-geometry alignment, with confidence intervals that do not overlap those of the other encoders. A layer-wise analysis further shows that masked reconstruction preserves this structure toward the output. On natural images, MAE represents surface color globally, while language-supervised models encode color more strongly in relation to the foreground object. These results show that human-like color grounding has several distinct aspects that should not be reduced to a single score.
Chinese Translation
视觉模型是否以人类的方式感知颜色？现有的颜色表示评估通常将其与几何空间（如 CIELAB）或离散颜色标签进行比较。这些参考捕捉了感知距离或类别成员资格，但并未反映人类组织颜色的渐进方式。我们针对一个模糊感知模型评估颜色基础，该模型包含86个与人类调查数据相匹配的渐进类别。该框架可以应用于任何图像编码器，并测量三个互补属性：类别边界、类别紧凑性和超越颜色几何的渐进对齐。在十一种视觉变换器编码器中，类别级结果大致相似，而渐进对齐则存在显著差异。掩码自编码器（Masked Autoencoders）实现了最强的超越几何对齐，其置信区间与其他编码器不重叠。逐层分析进一步表明，掩码重建在输出阶段保持了这一结构。在自然图像上，MAE（Masked Autoencoder）全球性地表示表面颜色，而语言监督模型则在与前景对象的关系中更强烈地编码颜色。这些结果表明，类人颜色基础具有多个独特方面，不应简化为单一评分。
cs.CV / 50 / 2607.13651
From Surface Forecasting to Observability Forecasting: A Latent World Model for Cloud-Aware EO Monitoring
从表面预测到可观测性预测：一种面向云的地球观测监测潜在世界模型
Abstract
The bottleneck of Earth Observation processing chains is not the arrival of new imagery but whether the surface is actually visible when the image arrives. We study this as an observability forecasting problem on EarthNet2021. Given recent multispectral imagery and exogenous weather drivers, the goal is to predict whether the next acquisition will be usable and, if not, when a usable view is likely to return. To do this, we adapt LeWorldModel, a joint-embedding predictive architecture world model, to cloud-aware Earth Observation sequences. The final pipeline converts raw minicubes into episodic HDF5 sequences with five image channels (blue, green, red, near-infrared, cloud mask) and eight meteorological and calendar covariates. The resulting model has 18.0M trainable parameters and is trained from scratch on 23,904 training episodes. The trained leWorldModel is evaluated under a locked protocol: linear probes are fit on train only, calibration choices are set on an internal validation split, and the fitted heads are then frozen for valsplit, IID, OOD, and extreme evaluation. On the full frozen-bundle observability benchmark, LeWorldModel consistently outperforms persistence. For next-step usability, balanced accuracy ranges from 0.769 to 0.887, compared with 0.493 to 0.556 for persistence. For exact first-usable-horizon prediction, accuracy ranges from 0.602 to 0.806, compared with 0.120 to 0.369 for persistence. Against a frozen LightGBM baseline fit on the same training windows, LeWorldModel is better on continuous clear/cloud regression and on exact recovery timing on valsplit, IID, and extreme, while LightGBM is stronger on the simpler binary any-usable-within-six task and is more robust on OOD. In separate sampled diagnostic analyses, LeWM also produces strong ranking-based anomaly signals under synthetic temporal inconsistencies.
Chinese Translation
地球观测处理链的瓶颈不在于新影像的到达，而在于当影像到达时表面是否实际可见。我们将此视为地球网2021（EarthNet2021）上的可观测性预测问题进行研究。给定近期的多光谱影像和外生天气驱动因素，目标是预测下一次获取的影像是否可用，如果不可用，则预测何时可能会返回可用视图。为此，我们将LeWorldModel（联合嵌入预测架构世界模型）调整为适应云感知的地球观测序列。最终的处理流程将原始小立方体转换为具有五个图像通道（蓝色、绿色、红色、近红外、云掩膜）和八个气象及日历协变量的情节HDF5序列。生成的模型具有1800万个可训练参数，并在23904个训练情节上从头开始训练。训练后的LeWorldModel在锁定协议下进行评估：线性探针仅在训练集上拟合，校准选择在内部验证分割上设置，然后冻结拟合的头部用于验证分割、独立同分布（IID）、超出分布（OOD）和极端评估。在完整的冻结包可观测性基准测试中，LeWorldModel始终优于持久性预测。对于下一步的可用性，平衡准确率范围从0.769到0.887，而持久性为0.493到0.556。对于精确的首次可用视野预测，准确率范围从0.602到0.806，而持久性为0.120到0.369。与在相同训练窗口上拟合的冻结LightGBM基线相比，LeWorldModel在连续的清晰/云回归和在验证分割、IID和极端情况下的精确恢复时机上表现更好，而LightGBM在更简单的二元“六小时内任何可用”任务上更强，并且在OOD上更具鲁棒性。在单独的采样诊断分析中，LeWM在合成时间不一致性下也产生了强大的基于排名的异常信号。
cs.CV / 51 / 2607.13653
Exploratory, Communicative, and Deployable: Vision-Driven Embodied Agents for Open-World Mobile Manipulation
探索性、交互性和可部署性：面向开放世界移动操作的视觉驱动体化智能体
Abstract
Real-world deployment of embodied agents requires active exploration, visual grounding, and interactive intent disambiguation. However, existing frameworks often rely on privileged simulator states or assume complete instructions, bypassing realistic deployment challenges. To bridge this gap, we present REAL, an agentic framework for open-world mobile manipulation. REAL establishes sim-to-real-consistent environment APIs without oracle perception and integrates a simulated user to enable human-in-the-loop interaction. Within this environment, we design diverse task compositions to drive data collection, supervised fine-tuning, and online reinforcement learning, systematically optimizing agent performance. To comprehensively evaluate this approach, we introduce REAL-Bench, a benchmark spanning 241 tasks across active exploration, visual distraction, articulated manipulation, and interactive disambiguation. Experimental results demonstrate that our trained agent outperforms leading commercial closed-source VLMs on interactive tasks with a 56.9% success rate. Further empirical analysis reveals that our hierarchical training pipeline successfully aligns the model's tool-use capabilities while maintaining robust open-vocabulary reasoning under extended exploration horizons. Finally, we deploy and evaluate our framework on a physical dual-arm mobile robot, where it achieves a 78.3% end-to-end success rate over 60 real-world episodes. These physical trials demonstrate robust zero-shot transferability to unseen household scenarios, validating that our sim-to-real-consistent design successfully bridges the reality gap for long-horizon mobile manipulation. Code is available at https://github.com/InternRobotics/REAL.
Chinese Translation
体化智能体在现实世界中的部署需要主动探索、视觉基础和交互意图消歧。然而，现有框架通常依赖于特权的模拟器状态或假设完整的指令，绕过了现实部署中的挑战。为了解决这一问题，我们提出了REAL，一个用于开放世界移动操作的智能体框架。REAL建立了无oracle感知的模拟到现实一致的环境API，并集成了一个模拟用户，以实现人机交互。在该环境中，我们设计了多样的任务组合，以推动数据收集、监督微调和在线强化学习，系统性地优化智能体性能。为了全面评估这一方法，我们引入了REAL-Bench，一个涵盖241个任务的基准，涉及主动探索、视觉干扰、关节操作和交互消歧。实验结果表明，我们训练的智能体在交互任务上以56.9%的成功率超越了领先的商业闭源视觉语言模型（VLM）。进一步的实证分析显示，我们的层次训练管道成功地对齐了模型的工具使用能力，同时在扩展探索范围内保持了强大的开放词汇推理能力。最后，我们在一个物理双臂移动机器人上部署并评估了我们的框架，在60个真实世界的实验中实现了78.3%的端到端成功率。这些物理试验展示了对未见家庭场景的强大零样本迁移能力，验证了我们的模拟到现实一致设计成功地弥合了长时间移动操作的现实差距。代码可在 https://github.com/InternRobotics/REAL 获取。
cs.CV / 52 / 2607.13654
T3HG-Editor: Text-driven 3D Human Garment Editing with Body Priors Embedded in SMPL-X
T3HG-Editor：基于文本驱动的3D人类服装编辑，嵌入SMPL-X的身体先验
Abstract
While 3D Gaussian Editing (3DGE) has seen substantial progress, text-driven 3D human garment editing remains largely underexplored. Existing 3DGE works typically follow a paradigm that applies 2D editing techniques to multi-view rendered images and updates 3D Gaussians based on the modified images. Extending such methods to 3D human garment editing suffers from low-fidelity outcomes, caused by introduced distortions and garment inconsistencies. A promising breakthrough opportunity arises from the SMPL eXpressive (SMPL-X) model that embodies rich prior information for virtual humans. Motivated by this insight, we propose a text-driven 3D human garment editor termed T3HG-Editor, which delivers high-fidelity and garment consistent results by leveraging geometry and joint priors embedded in SMPL-X. Specifically, T3HG-Editor contains three stages, namely obtainment of editable Gaussians, garment consistent editing, and Gaussian updating with overflow pruning. The obtainment of editable Gaussians begins with seeding Gaussians along SMPL-X normals to generate sufficient near surface Gaussians, followed by a 2D mask constraint that precisely localizes the target Gaussians to be edited. The garment consistent editing aggregates tokens corresponding to the same SMPL-X vertex across multiple views and propagates them to their original views, enforcing garment consistency without requiring additional training. Gaussian updating with overflow pruning employs a Signed Distance Function (SDF) defined on SMPL-X to construct a human distance field, which is then integrated with a 2D semantic mask to prune overflowing Gaussians, thus preventing contamination of non-target regions. Experiments on multiple subjects and diverse garment types demonstrate that T3HG-Editor outperforms state-of-the-art methods in both editing quality and garment consistency.
Chinese Translation
尽管3D高斯编辑（3DGE）已取得显著进展，但基于文本驱动的3D人类服装编辑仍然在很大程度上未得到探索。现有的3DGE研究通常遵循一种范式，即将2D编辑技术应用于多视角渲染图像，并根据修改后的图像更新3D高斯。将此类方法扩展到3D人类服装编辑时，由于引入的失真和服装不一致性，导致低保真度的结果。SMPL eXpressive（SMPL-X）模型提供了丰富的虚拟人类先验信息，为此带来了一个有希望的突破机会。基于这一见解，我们提出了一种名为T3HG-Editor的基于文本驱动的3D人类服装编辑器，通过利用嵌入在SMPL-X中的几何和关节先验，实现高保真度和服装一致性的结果。具体而言，T3HG-Editor包含三个阶段，即可编辑高斯的获取、服装一致性编辑和带溢出修剪的高斯更新。可编辑高斯的获取始于沿SMPL-X法线播种高斯，以生成足够的近表面高斯，随后通过2D掩模约束精确定位要编辑的目标高斯。服装一致性编辑聚合对应于多个视角中相同SMPL-X顶点的标记，并将其传播到原始视角，从而在不需要额外训练的情况下强制服装一致性。带溢出修剪的高斯更新利用在SMPL-X上定义的有符号距离函数（SDF）构建人类距离场，然后与2D语义掩模结合，以修剪溢出的高斯，从而防止非目标区域的污染。在多个对象和多样化服装类型上的实验表明，T3HG-Editor在编辑质量和服装一致性方面均优于最先进的方法。
cs.CV / 53 / 2607.13656
FreeLit: Paired-Free Indoor Relighting via Physics-Guided Diffusion
FreeLit：基于物理引导扩散的无配对室内重照明
Abstract
Image-based indoor scene relighting remains challenging due to the complex interplay between cluttered geometry and local illumination, requiring precise modeling of light position, color, and intensity. Existing data-driven methods implicitly learn this relationship via paired multi-illumination datasets. Nevertheless, this data is costly and fails to scale, which is essential for accurate light-source-level control. Conversely, inverse-rendering methods reduce the data dependency by incorporating physical priors; however, they lack the robustness of intrinsic estimation in challenging conditions. In this paper, we present FreeLit, a paired-free framework for controllable indoor relighting that explicitly manipulates light-source location, color, and intensity. Instead of relying on paired supervision, we construct a physics-guided illumination prior from intrinsic scene properties, generating a structured lightmap along with a pseudo-relit image to guide diffusion-based synthesis. To address instability in intrinsic estimation, especially in low-light scenes, we introduce a relighting-guided intrinsic stabilization strategy that enforces illumination-invariant reflectance through structure-aware distillation and consistency constraints. Furthermore, we propose controllability-oriented evaluation metrics to quantify alignment with user-specified illumination color and intensity. Experimental results demonstrate that FreeLit achieves stable, physically consistent, and controllable relighting, with improved robustness in low-light indoor scenes, without requiring paired supervision.
Chinese Translation
基于图像的室内场景重照明仍然面临挑战，因为杂乱几何与局部照明之间的复杂相互作用需要精确建模光源的位置、颜色和强度。现有的数据驱动方法通过配对的多光照数据集隐式学习这种关系。然而，这些数据成本高昂且难以扩展，而准确的光源级控制对此至关重要。相反，逆渲染方法通过结合物理先验减少了对数据的依赖；然而，在复杂条件下，它们缺乏内在估计的鲁棒性。本文提出了FreeLit，一个无配对的可控室内重照明框架，明确操控光源的位置、颜色和强度。我们并不依赖配对监督，而是从内在场景属性构建一个物理引导的照明先验，生成一个结构化的光照图以及一个伪重照明图像，以指导基于扩散的合成。为了解决内在估计的不稳定性，特别是在低光照场景中，我们引入了一种重照明引导的内在稳定化策略，通过结构感知蒸馏和一致性约束来强制照明不变的反射率。此外，我们提出了面向可控性的评估指标，以量化与用户指定的照明颜色和强度的一致性。实验结果表明，FreeLit实现了稳定、物理一致且可控的重照明，在低光照室内场景中表现出更好的鲁棒性，而无需配对监督。
cs.CV / 54 / 2607.13661
Fine-grained CLIP fine-tuning with self-annotated region alignment
基于自我标注区域对齐的细粒度 CLIP 微调
Abstract
Contrastive Language-Image Pre-training (CLIP) has been shown to have limitations in its fine-grained dense feature representation, due to its pre-training focusing on matching the whole image to a text description. Considering the large data and computational burden in pre-training a vision-language model from scratch, a series of works aim to enhance the fine-grained ability of CLIP through a fine-tuning scheme. However, existing works suffer from a variety of limitations: additional region annotations are usually required, which limits the semantic diversity due to the predefined categories and leads to a large effort to process the training data; and they usually sacrifice CLIP's original ability for global visual representation. To bypass these limitations, we propose SFF-CLIP (Self-annotated Fine-grained Fine-tuning for CLIP), which only uses image-text pairs as input to boost the fine-grained representation ability in the CLIP fine-tuning, while maintaining the global visual-semantic consistency. Concretely, a run-time region-phrase alignment scheme is designed, which obtains concept phrases from the input sentence, and aligns them with corresponding extracted region-based features using text-specific heat maps. Extensive experiments demonstrate that SFF-CLIP leads to significant performance improvements on fine-grained dense feature representation, as well as maintaining the performance of the original CLIP on image-level tasks. Code will be released later.
Chinese Translation
对比语言-图像预训练（CLIP）在细粒度密集特征表示方面存在局限性，这主要是由于其预训练过程集中于将整个图像与文本描述进行匹配。考虑到从头开始预训练视觉-语言模型所需的大量数据和计算负担，一系列研究旨在通过微调方案增强 CLIP 的细粒度能力。然而，现有工作存在多种局限性：通常需要额外的区域标注，这限制了语义多样性，因为预定义类别的存在导致处理训练数据的工作量巨大；而且它们通常牺牲了 CLIP 原有的全局视觉表示能力。为了绕过这些限制，我们提出了 SFF-CLIP（自我标注细粒度微调 CLIP），该方法仅使用图像-文本对作为输入，以提升 CLIP 微调中的细粒度表示能力，同时保持全局视觉-语义一致性。具体而言，我们设计了一种运行时区域-短语对齐方案，该方案从输入句子中获取概念短语，并使用特定于文本的热图将其与相应的提取区域特征对齐。大量实验表明，SFF-CLIP 在细粒度密集特征表示方面显著提高了性能，同时保持了原始 CLIP 在图像级任务上的表现。代码将在稍后发布。
cs.CV / 55 / 2607.13669
Learning Speaker Identity Beyond Language and Modality Constraints: Insights from the POLY-SIM 2026 Challenge
超越语言和模态限制的说话人身份学习：来自POLY-SIM 2026挑战的见解
Abstract
Multimodal speaker identification systems typically assume the availability of complete and homogeneous audio-visual modalities during both training and testing, and assume each speaker only speaks a single language. However, in real-world applications, such assumptions often do not hold. Visual or audio information may be missing due to occlusions, camera or microphone failures, or privacy constraints. Multilingual speakers introduce additional complexity due to linguistic variability across languages. These situations constitute substantial challenges for the robustness and generalization capabilities of multimodal speaker identification systems. Aim of the POLY-SIM 2026 challenge is to address these aspects of speaker identification and to provide a standardized setup for the comparison of the proposed solutions.
Chinese Translation
多模态说话人识别系统通常假设在训练和测试过程中可获得完整且同质的音频-视觉模态，并假设每位说话者仅使用单一语言。然而，在实际应用中，这些假设往往并不成立。由于遮挡、摄像头或麦克风故障，或隐私限制，视觉或音频信息可能会缺失。多语言说话者由于语言之间的变异性而引入了额外的复杂性。这些情况对多模态说话人识别系统的鲁棒性和泛化能力构成了重大挑战。POLY-SIM 2026挑战的目标是解决这些说话人识别方面的问题，并提供一个标准化的设置，以便比较所提出的解决方案。
cs.CV / 56 / 2607.13674
WAVE-Stereo: Warp-Aligned Volume Encoding for Stereo Matching
WAVE-Stereo：用于立体匹配的扭曲对齐体积编码
Abstract
Existing iterative stereo matching methods primarily adopt two types of correspondence representation: explicit matching search via correlation volumes and local residual refinement via warped features, yet the two remain separately modeled. We propose WAVE-Stereo, built on a core insight: correlation volumes and feature warping provide complementary matching cues. \textbf{GeoWarp Correspondence Encoder (GWCE)} encodes matching search, residual alignment, and disparity prior in parallel at the ConvGRU input. To mitigate matching degradation in textureless regions, we propose \textbf{Periodic Global Context Propagation (PGCP)}, which propagates global spatial information in a periodic manner. On five real-world benchmarks -- Middlebury, ETH3D, KITTI 2012, KITTI 2015, and Booster -- WAVE-Stereo achieves competitive zero-shot generalization accuracy without any external foundation model prior, achieving 3.18\% D1-all on KITTI 2015, 4.42\% Bad-2.0 on Booster, and 66ms real-time inference, striking a favorable balance between accuracy and efficiency. Our code is available at https://github.com/yamanoko-do/WAVE-Stereo.
Chinese Translation
现有的迭代立体匹配方法主要采用两种对应关系表示：通过相关体积进行显式匹配搜索，以及通过扭曲特征进行局部残差精细化，但这两者仍然是分开建模的。我们提出了WAVE-Stereo，基于一个核心见解：相关体积和特征扭曲提供了互补的匹配线索。\textbf{GeoWarp Correspondence Encoder (GWCE)}在ConvGRU输入中并行编码匹配搜索、残差对齐和视差先验。为了减轻在无纹理区域的匹配退化，我们提出了\textbf{Periodic Global Context Propagation (PGCP)}，它以周期性方式传播全局空间信息。在五个真实世界基准测试上——Middlebury、ETH3D、KITTI 2012、KITTI 2015和Booster——WAVE-Stereo在没有任何外部基础模型先验的情况下，实现了具有竞争力的零-shot泛化精度，在KITTI 2015上达到了3.18\% D1-all，在Booster上达到了4.42\% Bad-2.0，并实现了66毫秒的实时推理，在准确性和效率之间取得了良好的平衡。我们的代码可在https://github.com/yamanoko-do/WAVE-Stereo获取。
cs.CV / 57 / 2607.13681
Towards Spatial Supersensing in the Wild
迈向野外的空间超感知
Abstract
Humans can efficiently parse continuous sensory streams, from hours to years, scaffolding an internal world model that grounds spatial reasoning and prediction. To mimic this capacity, spatial supersensing challenges multimodal models to move beyond linguistic understanding toward true world modeling. However, their benchmark relies on synthetic long videos, formed by concatenating random short clips, and is mostly limited to household scenes, leaving real-world continuity and diversity underexplored. To address the gap, we introduce $\textbf{VSI-Super-Wild}$, a large-scale benchmark for evaluating spatial supersensing over long temporal horizons in diverse in-the-wild scenes. Notably, inspired by cognitive studies on how humans structure experience, we systematically probe the full triad of world state: the agent (observer), objects (scene items), and the environment (places and global layout). In total, VSI-Super-Wild contains $\textbf{6,980}$ human-verified question-answer pairs derived from $\textbf{442}$ real-world videos spanning 8 scene categories, including long-form recordings exceeding 4 hours. Results on VSI-Super-Wild expose a fundamental disconnect: despite advances in static image understanding, models consistently fail at tasks that require coherent world-state tracking over time. We characterize how performance degrades with world-state complexity and temporal horizon, and diagnose four failure modes: spatial collapse, semantic shortcuts, insufficient update, and instance confusion. This taxonomy reveals that models lack mechanisms to bind objects, agents, and environments into a unified spatial world model, a fundamental gap that defines the path forward for spatial supersensing.
Chinese Translation
人类能够高效地解析连续的感官流，从数小时到数年，构建一个内部世界模型，以支撑空间推理和预测。为了模拟这种能力，空间超感知挑战多模态模型超越语言理解，朝着真实世界建模的方向发展。然而，它们的基准测试依赖于合成的长视频，这些视频是通过连接随机短片段形成的，并且主要限于家庭场景，导致真实世界的连续性和多样性未得到充分探索。为了解决这一问题，我们引入了 $ extbf{VSI-Super-Wild}$，这是一个大规模基准，用于评估在多样化的真实场景中，跨越长时间跨度的空间超感知。值得注意的是，受到认知研究的启发，研究人类如何构建经验，我们系统地探讨了世界状态的完整三元组：代理（观察者）、物体（场景项目）和环境（地点和全球布局）。总的来说，VSI-Super-Wild 包含 $ extbf{6,980}$ 个经过人类验证的问题-答案对，源自 $ extbf{442}$ 个真实世界视频，涵盖8个场景类别，包括超过4小时的长时间录制。VSI-Super-Wild 的结果揭示了一个根本性的脱节：尽管静态图像理解取得了进展，但模型在需要随时间一致跟踪世界状态的任务中始终表现不佳。我们描述了性能如何随着世界状态复杂性和时间跨度的增加而下降，并诊断出四种失败模式：空间崩溃、语义捷径、不充分更新和实例混淆。这一分类法揭示了模型缺乏将物体、代理和环境结合成统一空间世界模型的机制，这是定义空间超感知未来发展的根本性缺口。
cs.CV / 58 / 2607.13682
Calibrated Closed-Form Uncertainty for Radiative Gaussian Splatting in Sparse-View CT
稀疏视图CT中辐射高斯溅射的校准闭式形式不确定性
Abstract
Radiative Gaussian splatting has made sparse-view CT reconstruction fast, but existing methods output point estimates with no notion of where the reconstruction can be trusted. We exploit a property of transmissive X-ray imaging that RGB splatting cannot claim -- projection and voxelization are strictly linear in the per-Gaussian densities -- to equip radiative Gaussians with a variational density posterior whose predictive variance propagates in closed form, exactly, in a single forward pass, in both volume space ($\sigma^2(x)=\sum_i g_i(x)^2 s_i^2$) and projection space ($\mathrm{Var}[I_p]=\sum_i w_{i,p}^2 s_i^2$). We present the first systematic calibration study for Gaussian-splatting CT (Spearman / AUSE / ECE with temperature scaling), showing that the resulting per-voxel uncertainty ranks true reconstruction error on 14 of 15 scenes of the official benchmark across three view budgets -- 9 of 15 additionally meeting our magnitude-calibration target after a single temperature -- while the perturbation-ensemble heuristic of concurrent work, transplanted to voxel space under the same protocol on our development scenes, does not (rank correlation as low as $-0.08$). We then dissect why uncalibrated acquisition scores can nevertheless select acceptable views, identifying three regimes -- flat (isotropic, balanced), pathological (degenerate coverage), and anisotropic -- and showing, in controlled single-scene testbeds, that principled uncertainty earns a measurable premium only in the last, motivating a coverage-gated, maturity-scheduled acquisition policy; the same calibrated posterior further points toward a dose-adaptive stopping rule, whose experimental validation we leave to future work.
Chinese Translation
辐射高斯溅射使得稀疏视图CT重建变得快速，但现有方法输出的仅为点估计，未能指明重建结果的可信度。我们利用透射X射线成像的一个特性，这是RGB溅射无法声称的——每个高斯密度的投影和体素化在严格意义上是线性的——为辐射高斯分布配备了一个变分密度后验，其预测方差在体积空间（$ ext{σ}^2(x)= ext{∑}_i g_i(x)^2 s_i^2$）和投影空间（$ ext{Var}[I_p]= ext{∑}_i w_{i,p}^2 s_i^2$）中以闭式形式精确传播，仅需一次前向传递。我们展示了高斯溅射CT的首个系统校准研究（Spearman / AUSE / ECE与温度缩放），结果表明，得到的每个体素不确定性在15个官方基准场景中的14个场景上能够有效排名真实重建误差——在三种视图预算下，有9个场景在经过一次温度调整后也满足我们的幅度校准目标——而同时进行的工作中的扰动集成启发式方法，在相同协议下移植到体素空间后并未达到这一效果（排名相关性低至$-0.08$）。随后，我们分析了为何未校准的采集分数仍然能够选择可接受的视图，识别出三种状态——平坦（各向同性，平衡）、病态（退化覆盖）和各向异性——并在控制的单场景测试平台上展示，原则性不确定性仅在最后一种状态下获得可测量的优势，这激励了一个基于覆盖的、成熟度调度的采集策略；同样的校准后验进一步指向了一种剂量自适应的停止规则，其实验验证留待未来工作。
cs.CV / 59 / 2607.13685
DNA: Dual-stage Native Attribution for Generated Image Source Tracing
DNA：生成图像源追踪的双阶段原生归属方法
Abstract
The rapid evolution of image generation has produced numerous within-family variants, making source-model attribution of suspect images increasingly important for digital forensics. Existing proactive methods rely on watermark embedding or model modification, which may degrade visual quality and limit deployment flexibility. Passive methods often rely on large-scale supervised training or a single reconstruction signal, limiting their ability to handle unknown sources and distinguish highly similar within-family variants. We observe that attribution signals in latent generative models are naturally stratified across architectural levels: VAE-level cues reflect family-shared information, whereas backbone-level cues capture variant-specific behaviors. Motivated by this insight, we propose Dual-stage Native Attribution (DNA), a coarse-to-fine framework that follows this hierarchy without additional neural-network training. The coarse-grained stage uses Autoencoder Double-Reconstruction (AEDR) for efficient open-set family-level screening. The fine-grained stage performs closed-set model-level attribution with Native Prediction Consistency (NPC), which compares native prediction errors of within-family variants across multiple noise levels under semantic conditioning and attributes the source via normalized calibrated scores. To enable systematic evaluation, we construct DNA-30K, a benchmark for within-family variant attribution under open-set family-level evaluation. It comprises 30,000 images generated by 24 candidate models across six families spanning both denoising diffusion and flow matching, plus non-candidate generated and natural images as unknown sources. Experiments show that DNA achieves 89.11% end-to-end attribution accuracy on a task where random guessing accuracy is below 1% and outperforms the strongest baseline by 33.81% even when AEDR is used as the coarse-grained stage.
Chinese Translation
图像生成的快速发展产生了众多同一家族的变体，使得可疑图像的源模型归属在数字取证中变得愈加重要。现有的主动方法依赖于水印嵌入或模型修改，这可能会降低视觉质量并限制部署灵活性。被动方法通常依赖于大规模的监督训练或单一重建信号，限制了其处理未知源和区分高度相似的同一家族变体的能力。我们观察到潜在生成模型中的归属信号在架构层面上自然分层：变分自编码器（VAE）级别的线索反映了家族共享的信息，而主干级别的线索则捕捉了变体特定的行为。基于这一见解，我们提出了双阶段原生归属（DNA）框架，这是一个从粗到细的框架，遵循这一层次结构而无需额外的神经网络训练。粗粒度阶段使用自编码器双重重建（AEDR）进行高效的开放集家族级筛选。细粒度阶段通过原生预测一致性（NPC）进行闭集模型级归属，该方法比较同一家族变体在语义条件下多个噪声水平下的原生预测误差，并通过归一化校准分数归属源。为了实现系统评估，我们构建了DNA-30K，这是一个用于开放集家族级评估的同一家族变体归属基准。它包含了由24个候选模型生成的30,000张图像，涵盖了六个家族，涉及去噪扩散和流匹配，以及作为未知源的非候选生成图像和自然图像。实验表明，DNA在一个随机猜测准确率低于1%的任务上实现了89.11%的端到端归属准确率，并且即使在使用AEDR作为粗粒度阶段时，仍比最强基线提高了33.81%。
cs.CV / 60 / 2607.13689
Barnamala: Parameter-Efficient Handwritten Devanagari Recognition at Benchmark Saturation
Barnamala：基于参数效率的手写天城文识别在基准饱和下的研究
Abstract
We built a compact convolutional network (1.11 M parameters) for 46-class DHCD Devanagari recognition and reached 99.73%, the highest reported at 15.6x smaller than prior state-of-the-art. We have effectively reached the saturation point: every model tested, large teacher ensembles included, hits the same 11-error intrinsic floor. No configuration achieves a statistically clear win under exact McNemar tests with Wilson confidence intervals. Even without knowledge distillation, our student matches the nearest large-model baseline (17.32 M parameters; McNemar $p = 0.345$). Outside of DHCD, zero-shot on CMATERdb digits gives 76.6% and fine-tuning reaches 97.8%; corruption robustness is also far better than large baselines (mean corruption accuracy 75.7% vs. 38.7%). All artifacts are at https://github.com/Ampixa/barnamala.
Chinese Translation
我们构建了一个紧凑的卷积网络（1.11 M 参数），用于46类 DHCD 天城文识别，达到了99.73%的准确率，这是迄今为止报告的最高水平，且模型大小比之前的最先进技术小15.6倍。我们有效地达到了饱和点：每个测试过的模型，包括大型教师集成，均达到了相同的11个错误内在底线。在精确的 McNemar 测试中，没有任何配置在威尔逊置信区间下取得统计显著的胜利。即使没有知识蒸馏，我们的学生模型也与最近的大型模型基线（17.32 M 参数；McNemar $p = 0.345$）相匹配。在 DHCD 之外，对 CMATERdb 数字的零样本测试得分为76.6%，而微调后达到了97.8%；在抗干扰能力方面也远超大型基线（平均抗干扰准确率75.7%对比38.7%）。所有相关材料可在 https://github.com/Ampixa/barnamala 获取。
cs.CV / 61 / 2607.13712
Groc-PO: Grounded Context Preference Optimization for Truthful Multimodal LLMs
Groc-PO：用于真实多模态大语言模型的基础上下文偏好优化
Abstract
Despite the rapid progress of Multimodal Large Language Models (MLLMs), they still suffer from untruthfulness issues, such as visual hallucinations, content fabrication, and unfaithful reasoning, which substantially undermine their faithfulness and practical utility. Alignment methods based on human preference, such as Direct Preference Optimization (DPO), have been widely adopted to address these issues. However, multimodal reasoning errors often propagate across stages, and final-answer errors can often be traced to mistakes in early grounding stages, yet standard DPO typically applies preference optimization at the final-answer level. This credit-assignment challenge means that supervision for early grounding stages is indirect rather than stage-specific, making it difficult to suppress error propagation arising from grounding drift and context inconsistency. To address this, we propose Grounded Context Preference Optimization (Groc-PO), a grounded preference optimization framework for MLLMs. We further construct the Grounded Context Preference Dataset (GCPD), organizing multi-stage preference samples around three stages of Object Grounding, Contextual Grounding, and Grounded Reasoning, to capture the formation, integration, and utilization of grounded context. By introducing more explicit preference supervision over multiple grounded stages, Groc-PO strengthens context-dependent reasoning and mitigates cross-stage error propagation. Extensive experiments show that, compared with standard DPO and other strong baselines, Groc-PO achieves improved performance in hallucination mitigation, faithful reasoning, and overall reliability, supporting the value of more explicit grounded supervision for trustworthy multimodal reasoning.
Chinese Translation
尽管多模态大语言模型（MLLMs）取得了快速进展，但它们仍然面临不真实的问题，例如视觉幻觉、内容虚构和不可靠推理，这些问题严重削弱了它们的可信度和实际效用。基于人类偏好的对齐方法，如直接偏好优化（DPO），已被广泛采用以解决这些问题。然而，多模态推理错误往往在各个阶段之间传播，最终答案的错误通常可以追溯到早期基础阶段的错误，而标准的DPO通常仅在最终答案层面进行偏好优化。这种信用分配挑战意味着对早期基础阶段的监督是间接的，而不是特定于阶段的，这使得抑制因基础漂移和上下文不一致而产生的错误传播变得困难。为了解决这个问题，我们提出了基础上下文偏好优化（Groc-PO），这是一个针对MLLMs的基础偏好优化框架。我们进一步构建了基础上下文偏好数据集（GCPD），围绕对象基础、上下文基础和基础推理三个阶段组织多阶段偏好样本，以捕捉基础上下文的形成、整合和利用。通过在多个基础阶段引入更明确的偏好监督，Groc-PO增强了基于上下文的推理，并减轻了跨阶段的错误传播。大量实验表明，与标准DPO和其他强基线相比，Groc-PO在幻觉缓解、可靠推理和整体可靠性方面取得了更好的性能，支持了对可信多模态推理进行更明确的基础监督的价值。
cs.CV / 62 / 2607.13738
Anatomically Faithful but Temporally Blind: Auditing Attribution for Left-Ventricular Ejection-Fraction Estimation from Echocardiography
解剖上忠实但时间上盲目：对超声心动图左心室射血分数估计的归因审计
Abstract
Background and Objective: Deep video models estimate left-ventricular ejection fraction (EF) from echocardiography with near-expert accuracy, and post-hoc attribution (Chefer relevance for transformers, Grad-CAM for CNNs) is increasingly used to certify that models "look at the right place." Yet whether these explanations are faithful both spatially and temporally is unaudited. Because EF is defined by the end-systolic (ES) and end-diastolic (ED) frames, a faithful explanation must localize the left ventricle (space) and the decisive frames (time). Methods: We fine-tune two distinct EF regressors on EchoNet-Dynamic -- a self-supervised VideoMAE transformer and a Kinetics-pretrained R(2+1)D CNN -- and audit each with architecture-matched attribution along three axes: intersection-over-relevance (IoR) against LV masks, deletion AUC, and a temporal localization index on ES/ED frames, each relative to chance with per-case 95% CIs over 50 studies. A tubelet-occlusion probe separates attribution failure from model behavior. Results: Both models are anatomically faithful -- IoR 2.91x (VideoMAE) and 1.98x (R(2+1)D) above chance -- yet temporally blind: temporal localization is indistinguishable from chance (0.97--1.00) and no better than random attribution. Occlusion shows the models do not preferentially rely on ES/ED (0.90x chance), so temporal blindness reflects model behavior, not an attribution artifact. Conclusions: Spatial faithfulness does not imply temporal faithfulness. Attribution can certify anatomical grounding while masking that a model ignores the clinically decisive frames -- a caution for XAI-based validation of video diagnostic models and a call for temporally-aware training and evaluation.
Chinese Translation
背景与目的：深度视频模型能够以接近专家的准确性从超声心动图中估计左心室射血分数（EF），并且后期归因（如变压器的 Chefer 相关性，卷积神经网络的 Grad-CAM）越来越多地被用来验证模型是否“关注正确的位置”。然而，这些解释在空间和时间上是否都忠实尚未经过审计。由于 EF 是由收缩末期（ES）和舒张末期（ED）帧定义的，因此一个忠实的解释必须定位左心室（空间）和决定性帧（时间）。方法：我们在 EchoNet-Dynamic 上微调了两个不同的 EF 回归模型——一个自监督的 VideoMAE 变压器和一个基于 Kinetics 预训练的 R(2+1)D 卷积神经网络，并沿三个轴对每个模型进行架构匹配的归因审计：与 LV 掩膜的相关性交集（IoR）、删除 AUC 和在 ES/ED 帧上的时间定位指数，每个相对于随机情况的 95% 置信区间（CI）基于 50 个研究。一个 tubelet-遮挡探针将归因失败与模型行为分开。结果：两个模型在解剖上都是忠实的——IoR 分别为 2.91 倍（VideoMAE）和 1.98 倍（R(2+1)D），高于随机情况——但在时间上是盲目的：时间定位与随机情况无异（0.97--1.00），并且没有比随机归因更好。遮挡显示模型并不优先依赖于 ES/ED（0.90 倍随机情况），因此时间盲目反映了模型行为，而不是归因伪影。结论：空间忠实性并不意味着时间忠实性。归因可以证明解剖基础，同时掩盖模型忽视临床决定性帧的事实——这是对基于可解释人工智能（XAI）的视频诊断模型验证的警示，也是对时间感知训练和评估的呼吁。
cs.CV / 63 / 2607.13792
EgoProceVQA: A Novel Egocentric Procedural Understanding Task with Self-Skill-Exploration Agent
EgoProceVQA：一种新颖的自我中心程序理解任务与自我技能探索智能体
Abstract
Most daily activities are inherently procedural. However, existing evaluations for egocentric video understanding seldom address procedural understanding and largely overlook complex key-step-level reasoning under the widely used video question answering (VQA) paradigm for MLLMs. Such capabilities are crucial for building procedural AI assistants deployable on wearable devices. To bridge this gap, we introduce the Egocentric Procedural Understanding VQA task (EgoProceVQA), which systematically evaluates egocentric procedural reasoning abilities of current MLLMs and agents through six types of key-step-centric questions. Furthermore, we develop EgoProceGen, a data generation platform that efficiently constructs QA data tailored to different question types. Based on this platform, we build a benchmark with 3,600 questions, four common procedural scenarios, and 31 everyday procedural tasks. Evaluations on EgoProceVQA show that existing MLLMs and agents still have substantial room for improvement in procedural understanding. Therefore, we further propose EgoProceAgent, a self-skill-exploration agentic framework. We design a generic tool library for procedural understanding and a standardized sub-skill library shared across tools and models, enabling self-exploration without ground-truth supervision. By exploring how to compose and select sub-skills, the agent discovers effective skill strategies for diverse problems, and attains state-of-the-art performance among open-source models on multiple tasks. Together, our benchmark, generation platform, and agentic framework establish a unified foundation for EgoProceVQA. Project page: https://z1oong.github.io/EgoProceVQA/.
Chinese Translation
大多数日常活动本质上是程序性的。然而，现有的自我中心视频理解评估很少涉及程序理解，并且在广泛使用的视频问答（VQA）范式下，往往忽视了复杂的关键步骤级推理。这些能力对于构建可部署在可穿戴设备上的程序性人工智能助手至关重要。为了解决这一问题，我们提出了自我中心程序理解VQA任务（EgoProceVQA），该任务通过六种类型的关键步骤中心问题系统地评估当前多模态大语言模型（MLLMs）和智能体的自我中心程序推理能力。此外，我们开发了EgoProceGen，一个数据生成平台，能够高效构建针对不同问题类型的问答数据。基于该平台，我们建立了一个基准，包含3600个问题、四种常见程序场景和31个日常程序任务。对EgoProceVQA的评估表明，现有的MLLMs和智能体在程序理解方面仍有很大的改进空间。因此，我们进一步提出了EgoProceAgent，一个自我技能探索的智能体框架。我们设计了一个通用的程序理解工具库和一个在工具和模型之间共享的标准化子技能库，使得在没有真实监督的情况下进行自我探索成为可能。通过探索如何组合和选择子技能，智能体发现了针对多种问题的有效技能策略，并在多个任务上达到了开源模型中的最先进性能。我们的基准、生成平台和智能体框架共同为EgoProceVQA建立了统一的基础。项目页面：https://z1oong.github.io/EgoProceVQA/
cs.CV / 64 / 2607.13800
Prospective clinical indication, post-hoc report leakage, and fusion design in multi-image chest radiograph classification: a patient-clustered evaluation
多图像胸部X光分类中的前瞻性临床指征、事后报告泄漏和融合设计：患者聚类评估
Abstract
Chest radiograph datasets often combine multiple images with Clinical Indication, Findings, and Impression, although these inputs are produced at different stages of care. We evaluated 15,000 ReXGradient-160K studies with two readable images and five CheXbert-derived report observations. Frozen DenseNet-121 and Bio+ClinicalBERT encoders were used to compare image-only, Indication-only, fixed-order multimodal, random-swap, DeepSets, and SectionGuard-MI models. Findings and Impression were evaluated only as post-hoc leakage controls. Models were trained with five seeds, and public-test uncertainty was estimated with 2,000 patient-cluster bootstrap replicates. Under U-Ones, macro AUROC was 0.643 for the primary image, 0.694 for two images, 0.749 for Indication, and 0.780 for ordinary two-image-plus-Indication fusion. SectionGuard-MI achieved AUROC 0.783 and AUPRC 0.260. Relative to ordinary fusion, its paired AUROC difference was 0.0031 (95% CI, -0.0042 to 0.0104; adjusted p=0.374), while its AUPRC difference was 0.0289 (95% CI, 0.0095 to 0.0413; adjusted p=0.004). DeepSets had the highest prospective AUROC point estimate (0.787), and random-swap fusion had the highest prospective AUPRC point estimate (0.265) with better calibration than SectionGuard-MI. Full report text alone reached AUROC 0.979 and AUPRC 0.836; AUROC remained above 0.973 after exact or expanded masking. These results show that prospective Indication is strongly associated with report-derived targets, permutation-aware fusion is competitive, and post-hoc report text creates substantial report-label circularity.
Chinese Translation
胸部X光数据集通常结合多个图像与临床指征、发现和印象，尽管这些输入是在不同的护理阶段产生的。我们评估了15,000个ReXGradient-160K研究，包含两个可读图像和五个基于CheXbert的报告观察。使用冻结的DenseNet-121和Bio+ClinicalBERT编码器比较了仅图像、仅指征、固定顺序多模态、随机交换、DeepSets和SectionGuard-MI模型。发现和印象仅作为事后泄漏控制进行评估。模型使用五个种子进行训练，并通过2,000个患者聚类自助法重复估计公共测试的不确定性。在U-Ones下，主要图像的宏观AUROC为0.643，两个图像为0.694，指征为0.749，普通的两个图像加指征融合为0.780。SectionGuard-MI达到了AUROC 0.783和AUPRC 0.260。与普通融合相比，其配对AUROC差异为0.0031（95% CI，-0.0042至0.0104；调整后p=0.374），而其AUPRC差异为0.0289（95% CI，0.0095至0.0413；调整后p=0.004）。DeepSets具有最高的前瞻性AUROC点估计（0.787），随机交换融合具有最高的前瞻性AUPRC点估计（0.265），且其校准优于SectionGuard-MI。仅完整报告文本达到了AUROC 0.979和AUPRC 0.836；在精确或扩展掩蔽后，AUROC仍保持在0.973以上。这些结果表明，前瞻性指征与报告衍生目标密切相关，考虑置换的融合具有竞争力，而事后报告文本造成了显著的报告标签循环性。
cs.CV / 65 / 2607.13802
RainDancer: RGB-Event Video Deraining with Rain-Oriented Spiking Dynamics
RainDancer：基于雨水导向脉冲动态的RGB-事件视频去雨方法
Abstract
Video deraining aims to recover clean visual content from rainy videos for reliable perception under adverse weather. Existing methods mainly rely on RGB sequences and temporal redundancy, but RGB-only restoration remains ambiguous in dynamic rainy scenes, where rain streaks, textures, boundaries, motion, and occlusions may share similar visual patterns. Event cameras provide complementary motion-sensitive cues with high temporal resolution, but event streams also contain sensor noise and background-triggered responses, so direct RGB-Event fusion may introduce cross-modal interference. To address this issue, we propose RainDancer, a progressive RGB-Event video deraining framework based on a decompose-before-interact paradigm. The core idea is to separate rain and background components within each modality before cross-modal interaction. In the RGB branch, frame features are progressively decomposed into rain and background representations. In the event branch, a rain-oriented spiking neural network module captures sparse and bursty event dynamics associated with rain motion. Component-level fusion is then performed between semantically aligned representations for structure preservation and rain suppression. We further introduce event-domain supervision to regularize sparse event reconstruction, structural consistency, and gradient orientation. Experiments on synthetic and real RGB-Event video deraining datasets demonstrate superior quantitative performance, visual quality, and downstream perception robustness. Code is available at https://github.com/AE86-plus/RainDancer.
Chinese Translation
视频去雨旨在从雨天视频中恢复干净的视觉内容，以便在恶劣天气条件下实现可靠的感知。现有方法主要依赖RGB序列和时间冗余，但仅依靠RGB的恢复在动态雨天场景中仍然模糊，因为雨滴、纹理、边界、运动和遮挡可能共享相似的视觉模式。事件相机提供了高时间分辨率的运动敏感线索，但事件流也包含传感器噪声和背景触发的响应，因此直接进行RGB-事件融合可能会引入跨模态干扰。为了解决这个问题，我们提出了RainDancer，一个基于分解后交互范式的渐进式RGB-事件视频去雨框架。其核心思想是在跨模态交互之前，先在每个模态内分离雨水和背景成分。在RGB分支中，帧特征逐步分解为雨水和背景表示。在事件分支中，一个雨水导向的脉冲神经网络模块捕捉与雨水运动相关的稀疏和突发事件动态。然后在语义对齐的表示之间进行成分级融合，以保持结构并抑制雨水。我们进一步引入事件域监督，以规范稀疏事件重建、结构一致性和梯度方向。对合成和真实RGB-事件视频去雨数据集的实验表明，所提方法在定量性能、视觉质量和下游感知鲁棒性方面表现优越。代码可在 https://github.com/AE86-plus/RainDancer 获取。
cs.CV / 66 / 2607.13805
AspectCLIP: Optimizing CLIP Representation Space via Aspect-Guided Consistency Regularization
AspectCLIP：通过方面引导的一致性正则化优化 CLIP 表示空间
Abstract
Contrastive Language-Image Pretraining learns a shared representation space through large-scale contrastive learning. However, existing methods that enforce global consistency regularization overlook a key challenge: the inherent information asymmetry between images and text: captions typically describe only one specific aspect of an image, thus images with similar visual content can be paired with completely divergent textual content and semantic information. Consequently, global regularizers inadvertently impose constraints between visually similar images whose captions describe divergent aspects, introducing semantic distortion into the representation space. We propose AspectCLIP, a framework that reformulates consistency regularization to respect this one-to-many structure. AspectCLIP first partitions training samples into attribute clusters based on textual similarity to identify aspect-coherent groups, then applies full cyclic consistency within each cluster while restricting cross-cluster regularization to prototype-level comparisons. This aspect-guided regularization enforces strict geometric alignment only when images and texts describe a consistent facet, while allowing flexibility across divergent aspects. Extensive experiments on downstream tasks demonstrate that AspectCLIP consistently outperforms traditional methods and achieves a more structured representation space.
Chinese Translation
对比语言-图像预训练通过大规模对比学习学习共享表示空间。然而，现有的强制全局一致性正则化的方法忽视了一个关键挑战：图像和文本之间固有的信息不对称性：标题通常只描述图像的一个特定方面，因此具有相似视觉内容的图像可以与完全不同的文本内容和语义信息配对。因此，全局正则化器无意中对视觉上相似但标题描述不同方面的图像施加约束，从而在表示空间中引入语义扭曲。我们提出了 AspectCLIP，一个重新构建一致性正则化以尊重这种一对多结构的框架。AspectCLIP 首先根据文本相似性将训练样本划分为属性聚类，以识别方面一致的组，然后在每个聚类内应用全循环一致性，同时将跨聚类正则化限制为原型级比较。这种方面引导的正则化仅在图像和文本描述一致的方面时强制严格的几何对齐，同时允许在不同方面之间的灵活性。在下游任务上的大量实验表明，AspectCLIP 一直优于传统方法，并实现了更结构化的表示空间。
cs.CV / 67 / 2607.13808
Bake It Till You Make It: Ultrafast Spatial Texture-Atlas Splatting
坚持到底：超快速空间纹理图集喷涂
Abstract
Recent extensions of 3D Gaussian Splatting (3DGS) capture fine color details using hash-grid-based appearance parameterization but incur high computational cost during fragment rendering. We introduce a decoupled radiance representation that models low-frequency geometry and view dependent appearance features with 2D surfels while representing high-frequency textures via a view-independent spatial hash grid that is baked into a compact texture atlas. By including sparsity-enhancing optimizations that penalize semi-transparency and per-primitive falloff, our method aggressively prunes insignificant surfels and achieves significantly faster and sparser reconstructions than prior work. Exploiting geometric sparsity and efficient GPU texture mapping, our approach achieves up to a fivefold speedup over 3DGS while preserving state-of-the-art visual fidelity, enabling real-time 4K rendering at 60 FPS on consumer hardware.
Chinese Translation
最近对3D高斯喷涂（3DGS）的扩展通过基于哈希网格的外观参数化捕捉细腻的颜色细节，但在片段渲染过程中会产生高计算成本。我们引入了一种解耦的辐射表示法，该方法使用2D表面（surfels）建模低频几何和视角依赖的外观特征，同时通过一个独立于视角的空间哈希网格表示高频纹理，并将其烘焙成一个紧凑的纹理图集。通过包含惩罚半透明性和每个原始体衰减的稀疏性增强优化，我们的方法积极修剪不重要的表面，达到了比之前的工作显著更快且更稀疏的重建效果。利用几何稀疏性和高效的GPU纹理映射，我们的方法在保持最先进的视觉保真度的同时，相较于3DGS实现了最高五倍的速度提升，使得在消费级硬件上实现实时4K渲染达到60帧每秒成为可能。
cs.CV / 68 / 2607.13826
Multimodal Assessment of Pancreatic Cancer Resectability Using Deep Learning
基于深度学习的胰腺癌切除可行性多模态评估
Abstract
Accurate determination of pancreatic ductal adenocarcinoma (PDAC) resectability relies on evaluating how the tumor interacts with major peripancreatic vessels on CT imaging, yet expert assessment often shows substantial variability. We introduce a fully automated multimodal deep learning framework that jointly analyzes 3D contrast enhanced CT and structured clinical information to classify patients into the three National Comprehensive Cancer Network (NCCN) resectability categories (upfront resectable, borderline resectable, locally advanced). The approach uses a Swin-UNETR backbone to obtain anatomy aware image representations through auxiliary segmentation of pancreas, tumor, and vascular structures. These features are fused with a compact clinical embedding derived from 17 routinely collected variables and processed by a lightweight classification head. Model training is guided by a dynamic multitask objective that adapts the balance between segmentation and classification based on current tumor Dice performance, promoting feature representations that remain both anatomically informed and discriminative.
Chinese Translation
准确判断胰腺导管腺癌（PDAC）的切除可行性依赖于评估肿瘤与主要胰腺周围血管在CT影像上的相互作用，但专家评估往往显示出显著的变异性。我们提出了一种完全自动化的多模态深度学习框架，该框架联合分析三维对比增强CT图像和结构化临床信息，以将患者分类为三种国家综合癌症网络（NCCN）切除可行性类别（可立即切除、边缘可切除、局部晚期）。该方法采用Swin-UNETR骨干网络，通过对胰腺、肿瘤和血管结构的辅助分割获取解剖学相关的图像表示。这些特征与从17个常规收集变量中提取的紧凑临床嵌入进行融合，并通过轻量级分类头进行处理。模型训练由动态多任务目标引导，该目标根据当前肿瘤Dice性能调整分割和分类之间的平衡，促进保持解剖学信息和区分性的特征表示。
cs.CV / 69 / 2607.13860
Towards Enhancing 3D Spatial Reasoning in Medical Multimodal Large Language Models
提升医学多模态大语言模型中的三维空间推理能力
Abstract
While Multimodal Large Language Models (MLLMs) have demonstrated remarkable success in 2D medical image understanding, their extension to 3D volumetric imaging remains hindered by prohibitive annotation costs and dataset opacity. Current data formats, predominantly consisting of rigid Visual Question Answering (VQA) pairs or unstructured final clinical reports, typically fail to capture explicit clinical reasoning. To address this limitation, we introduce a large-scale structured reasoning dataset constructed via a novel slice-wise data synthesis paradigm. Inspired by the genuine diagnostic workflow of radiologists, this paradigm models visual cognition by decomposing the complex 3D reading process, translating global clinical priors into fine-grained, per-slice observations that are subsequently synthesized into an interpretable Chain-of-Thought (CoT). Crucially, this synthesized reasoning framework enforces essential clinical principles: sequential spatial tracking, multi-slice spatial awareness for artifact mitigation, and differential exclusion. To validate this approach, we instruction-tune a standard 2D-pretrained MLLM baseline using the synthesized data to enhance its volumetric comprehension. Comprehensive evaluations across multiple 3D medical benchmarks demonstrate that our method yields significant performance improvements over the 2D baseline. Furthermore, the resulting model exhibits robust spatial reasoning capabilities and rivals resource-intensive native 3D architectures, effectively bridging the performance gap. Ultimately, this data-centric strategy unlocks deep volumetric understanding and highly interpretable clinical logic without requiring computationally expensive 3D-specific pre-training. The complete repository, including datasets and training workflows, is publicly available at https://github.com/2020420145009/hounsfield.
Chinese Translation
尽管多模态大语言模型（MLLMs）在二维医学图像理解方面取得了显著成功，但其在三维体积成像中的应用仍受到高昂标注成本和数据集不透明性的限制。目前的数据格式主要由刚性的视觉问答（VQA）对或非结构化的最终临床报告组成，通常无法捕捉到明确的临床推理。为了解决这一局限性，我们引入了一个通过新颖的切片式数据合成范式构建的大规模结构化推理数据集。该范式受到放射科医生真实诊断工作流程的启发，通过将复杂的三维阅读过程分解，建模视觉认知，将全局临床先验转化为细粒度的逐切片观察，随后合成成可解释的思维链（Chain-of-Thought, CoT）。重要的是，这一合成推理框架强制执行基本的临床原则：顺序空间跟踪、多切片空间意识以减轻伪影影响以及差异排除。为了验证这一方法，我们使用合成数据对标准的二维预训练MLLM基线进行指令调优，以增强其体积理解能力。在多个三维医学基准测试中的全面评估表明，我们的方法在性能上显著优于二维基线。此外，所得到的模型展现出强大的空间推理能力，并与资源密集型的原生三维架构相媲美，有效缩小了性能差距。最终，这一以数据为中心的策略在无需计算开销巨大的三维特定预训练的情况下，解锁了深层体积理解和高度可解释的临床逻辑。完整的代码库，包括数据集和训练工作流程，已公开发布在 https://github.com/2020420145009/hounsfield。
cs.CV / 70 / 2607.13881
Unleashing Multimodal Large Language Models for Training-free HOI Detection in the Wild
释放多模态大型语言模型以实现无训练的人物-物体交互检测
Abstract
Human-object interaction detection (HOID) has traditionally been formulated as a supervised detection problem over predefined interaction categories. While such paradigms achieve strong performance on closed-set benchmarks, they fundamentally entangle interaction understanding with dataset-specific supervision, limiting their ability to generalize to open-world and compositional scenarios. Recent HOI detectors attempt to leverage MLLMs through prompting strategies to transfer interaction-specific knowledge. However, such prompt-based approaches primarily focus on extracting discriminative representations from pretrained models, while underexploring their inherent multimodal reasoning capabilities. As a result, they struggle to provide informative contextual reasoning for ambiguous and open-world interaction scenarios. In this work, we present AgentHOI, a training-free, agentic framework that transfers the generalist multimodal reasoning capabilities of foundation models to HOI detection in the wild. Instead of learning interaction classifiers, AgentHOI modularly orchestrates complementary vision foundation modules to perform open-ended semantic reasoning and spatial grounding in a coordinated manner. To address the challenges of incomplete interaction discovery and ambiguous localization in complex scenes, we introduce two key mechanisms: (1) Context-aware Multi-round Reasoning, which progressively refines interaction hypotheses to ensure exhaustive and compositional HOI discovery, and (2) Multifaceted Interaction Localization, which enhances grounding precision by generating instance-specific descriptions that integrate semantic, spatial, and appearance cues. Extensive experiments demonstrate that AgentHOI achieves superior performance over state-of-the-art supervised and weakly supervised methods in real-world settings, despite requiring no HOID data for training.
Chinese Translation
人物-物体交互检测（HOID）传统上被视为一个在预定义交互类别上进行的监督检测问题。尽管这种范式在封闭集基准测试中取得了良好的性能，但它们在本质上将交互理解与特定数据集的监督纠缠在一起，限制了其在开放世界和组合场景中的泛化能力。近期的人物-物体交互检测器尝试通过提示策略利用多模态大型语言模型（MLLMs）来转移交互特定知识。然而，这种基于提示的方法主要集中在从预训练模型中提取区分性表示，而对其固有的多模态推理能力探索不足。因此，它们在为模糊和开放世界交互场景提供信息丰富的上下文推理方面面临挑战。在本研究中，我们提出了AgentHOI，一个无训练的、具有代理性的框架，将基础模型的通用多模态推理能力转移到开放环境中的HOI检测上。AgentHOI并不是学习交互分类器，而是模块化地协调互补的视觉基础模块，以协同方式执行开放式的语义推理和空间定位。为了解决复杂场景中不完整交互发现和模糊定位的挑战，我们引入了两个关键机制：（1）上下文感知的多轮推理，逐步细化交互假设，以确保全面和组合的HOI发现；（2）多面交互定位，通过生成集成语义、空间和外观线索的实例特定描述来增强定位精度。大量实验表明，尽管AgentHOI在训练中不需要HOID数据，但在现实世界设置中仍优于最先进的监督和弱监督方法。
cs.CV / 71 / 2607.13892
Fine-Grained Vision-Language Pretraining with Organ-Conditioned Pattern Tokens for CT Understanding
基于器官条件模式标记的细粒度视觉-语言预训练用于CT理解
Abstract
Computed tomography (CT) vision-language pretraining from paired volumes and radiology reports is a scalable yet challenging task. Existing methods commonly adopt global scan-report contrast, which is scalable but obscures heterogeneous organ evidence. Meanwhile, direct organ-level alignment remains coarse, since the same anatomy can exhibit multiple distinct radiological appearances. Therefore, pretraining requires a finer alignment unit: the organ-conditioned radiological pattern. In this work, we propose OCP-CT, an organ-conditioned pattern-token alignment framework for CT vision-language pretraining. Specifically, OCP-CT preserves a stable global CT-report contrastive branch and introduces an organ pattern interface: sparse Mixture-of-Experts (MoE) routes image and text tokens according to latent radiological patterns, learnable slots query the routed tokens into continuous pattern tokens, and paired token contrast aligns image-text pattern tokens with structured soft targets built from report-derived clinical similarity. On the publicly available CT-RATE and RAD-ChestCT benchmarks, OCP-CT achieves average AUROCs of 84.5% and 69.9% for zero-shot abnormality diagnosis, respectively. Compared with the strongest prior reported results, these results yield absolute AUROC gains of 6.7 and 0.8 percentage points.
Chinese Translation
从配对的体积和放射学报告中进行计算机断层扫描（CT）视觉-语言预训练是一项可扩展但具有挑战性的任务。现有方法通常采用全局扫描-报告对比，这虽然可扩展，但会掩盖异质器官证据。同时，直接的器官级对齐仍然较为粗糙，因为相同的解剖结构可能表现出多种不同的放射学外观。因此，预训练需要一个更精细的对齐单元：器官条件放射学模式。在本研究中，我们提出了OCP-CT，一个用于CT视觉-语言预训练的器官条件模式标记对齐框架。具体而言，OCP-CT保留了一个稳定的全局CT-报告对比分支，并引入了一个器官模式接口：稀疏专家混合（Mixture-of-Experts, MoE）根据潜在的放射学模式路由图像和文本标记，可学习的槽位查询将路由的标记转化为连续的模式标记，配对标记对比将图像-文本模式标记与从报告派生的临床相似性构建的结构化软目标对齐。在公开可用的CT-RATE和RAD-ChestCT基准上，OCP-CT在零样本异常诊断中分别达到了84.5%和69.9%的平均AUROC。与之前报告的最强结果相比，这些结果分别带来了6.7和0.8个百分点的绝对AUROC提升。
cs.CV / 72 / 2607.13905
The 2nd International StepUP Competition for Biometric Footstep Recognition: From Steps to Strides
第二届国际步态识别比赛：从脚步到步幅
Abstract
The International StepUP Competition Series was launched to advance research in pressure-based footstep biometrics through a standardized and challenging evaluation framework. Using the large-scale StepUP-P150 dataset (with more than 200,000 high-resolution dynamic footsteps from 150 individuals) and a previously unreleased test set, the 2nd edition of the competition addressed three key challenges: (1) generalization to unseen users with limited enrollment data, (2) robustness to domain shift caused by variations in footwear and walking speed and (3) effective fusion of paired left-right footsteps. While the first two challenges built on the inaugural competition, this edition introduced more extreme cross-domain conditions and moved beyond isolated footsteps to stride-level verification, enabling new opportunities for representation learning and inter-step information fusion. The competition attracted 26 registrants from academia and industry, with a best equal error rate of 8.00% achieved by the ArogyaPandit Research Team using a spatiotemporal CNN combined with an ensemble-based scoring strategy. The top solutions showcase the value of harnessing temporal patterns and of incorporating inference-time normalization and calibration strategies to improve scoring. However, the results also reveal that recognizing users in unseen personal footwear remains a challenge, especially in the presence of distractors with similar characteristics.
Chinese Translation
国际步态识别比赛系列旨在通过标准化和具有挑战性的评估框架，推动基于压力的脚步生物识别研究。利用大规模的StepUP-P150数据集（包含来自150个个体的超过200,000个高分辨率动态脚步）和一个之前未发布的测试集，第二届比赛解决了三个关键挑战：（1）对有限注册数据的未见用户的泛化能力，（2）对因鞋类和步行速度变化引起的领域转移的鲁棒性，以及（3）有效融合成对的左右脚步。虽然前两个挑战基于首届比赛的基础，但本届比赛引入了更极端的跨领域条件，并超越了孤立脚步，转向步幅级别的验证，从而为表示学习和步间信息融合提供了新的机会。比赛吸引了来自学术界和工业界的26名注册者，ArogyaPandit研究团队使用时空卷积神经网络（spatiotemporal CNN）结合基于集成的评分策略，达到了最佳等错误率8.00%。顶尖解决方案展示了利用时间模式以及在推理时进行归一化和校准策略以提高评分的价值。然而，结果也揭示了在未见的个人鞋类中识别用户仍然是一个挑战，尤其是在存在具有相似特征的干扰物时。
cs.CV / 73 / 2607.13925
Thresholded Cross-Attention for Reliable Intensity-Chromaticity Fusion in Low-Light Image Enhancement
阈值交叉注意力在低光图像增强中的可靠强度-色度融合
Abstract
Low-Light Image Enhancement (LLIE) requires a careful balance among noise suppression, color fidelity, and efficiency. Recent HVI-based methods alleviate color entanglement by decoupling intensity and chromaticity, yet how reliably the two streams are fused again is an overlooked factor that largely determines the final quality. We observe that the confidence of cross-stream attention is strongly layer-dependent, so the fixed-quota selection of Top-K sparse attention is mismatched to it, discarding informative dependencies in some layers while retaining noisy ones in others. Motivated by this observation, we propose TCA-Net, a network built around Thresholded Cross-Attention that targets reliable intensity-chromaticity fusion in the HVI space rather than introducing yet another color representation. At its core, TCA replaces the rigid Top-K quota with a fixed confidence threshold whose retained cardinality is input- and layer-adaptive, retaining only high-confidence cross-stream interactions while suppressing unreliable ones. Around this core, two complementary designs clean up the fusion before and after it: a Phase-guided Fourier Interaction Module provides a structure-aware brightness initialization for the intensity stream prior to fusion, and a Decoupled Dual-Stream Guidance Module constructs residual intensity features to suppress chromaticity leakage during reconstruction. A Scale-Aware Consistency Regularization further improves structural robustness under scale perturbations during training. Extensive experiments on LOL-v1, LOL-v2, Sony-Total-Dark, and LSRW-Huawei demonstrate that TCA-Net delivers competitive restoration accuracy, improved color fidelity, and a compact parameter size.
Chinese Translation
低光图像增强（LLIE）需要在噪声抑制、色彩保真度和效率之间进行仔细平衡。最近的基于HVI（高动态范围图像）的方法通过解耦强度和色度来缓解色彩纠缠，但如何可靠地再次融合这两个流是一个被忽视的因素，它在很大程度上决定了最终质量。我们观察到，跨流注意力的信心在很大程度上依赖于层，因此固定配额的Top-K稀疏注意力与之不匹配，导致某些层中丢弃了有信息的依赖关系，而在其他层中保留了噪声依赖关系。基于这一观察，我们提出了TCA-Net，一个围绕阈值交叉注意力构建的网络，旨在HVI空间中实现可靠的强度-色度融合，而不是引入另一种颜色表示。TCA的核心是用一个固定的信心阈值替换刚性的Top-K配额，其保留的基数是输入和层自适应的，仅保留高信心的跨流交互，同时抑制不可靠的交互。在这个核心周围，两个互补设计在融合前后清理融合过程：一个相位引导的傅里叶交互模块在融合之前为强度流提供结构感知的亮度初始化，而一个解耦双流引导模块构建残差强度特征，以抑制重建过程中色度泄漏。一个尺度感知一致性正则化进一步提高了训练过程中在尺度扰动下的结构鲁棒性。在LOL-v1、LOL-v2、Sony-Total-Dark和LSRW-Huawei上的大量实验表明，TCA-Net提供了竞争性的恢复精度、改善的色彩保真度和紧凑的参数规模。
cs.CV / 74 / 2607.13927
Cyclone: Diffusion Model for Cycle-Consistent Weather Editing from Unpaired Driving Data
Cyclone：基于无配对驾驶数据的循环一致天气编辑扩散模型
Abstract
Reliable perception under diverse weather conditions remains a major challenge for autonomous driving systems. A common strategy to improve robustness is either to synthesize adverse weather conditions for training perception models or to apply weather-removal techniques to recover clean inputs. However, existing approaches typically rely on synthetic data augmentation or physics-based, task-specific models that require paired training data and often struggle to generate realistic weather effects or generalize robustly to out-of-domain scenarios. Toward this problem, we present Cyclone, a unified framework for weather editing based on latent diffusion, equipped with cycle-consistent constraints and knowledge from image-text models. Cyclone enables the generation of multiple weather conditions across diverse scenes while eliminating the need for paired data. Experimental results show that our approach produces more realistic, structure-preserving outputs than existing baselines and leads to consistent improvements across several downstream driving perception tasks. Furthermore, we demonstrate that Cyclone can be distilled to a video diffusion model for temporally consistent weather editing.
Chinese Translation
在多样化天气条件下可靠感知仍然是自动驾驶系统面临的主要挑战。提高鲁棒性的常见策略是合成恶劣天气条件以训练感知模型，或应用天气去除技术以恢复干净输入。然而，现有方法通常依赖于合成数据增强或基于物理的、任务特定的模型，这些模型需要配对的训练数据，并且往往难以生成真实的天气效果或在域外场景中稳健地泛化。针对这一问题，我们提出了Cyclone，一个基于潜在扩散的统一天气编辑框架，配备循环一致性约束和来自图像-文本模型的知识。Cyclone能够在多样场景中生成多种天气条件，同时消除了对配对数据的需求。实验结果表明，我们的方法生成的输出比现有基线更具真实感且保留结构，并在多个下游驾驶感知任务中实现了一致的改进。此外，我们展示了Cyclone可以被提炼为一个视频扩散模型，以实现时间一致的天气编辑。
cs.CV / 75 / 2607.13931
SIVA-RL: Sensitivity-Invariance Visual Alignment for Multimodal Reinforcement Learning
SIVA-RL：用于多模态强化学习的敏感性不变视觉对齐
Abstract
Reinforcement learning with verifiable rewards (RLVR) drives multimodal reasoning, but answer-level correctness does not guarantee that a vision-language model grounds its predictions in visual evidence. Existing visual-intervention methods contrast policy behavior on original and modified images, yet assign supervision by the type of intervention rather than its observed effect. This assumption fails: identical operators produce heterogeneous outcomes across samples. We propose SIVA-RL, a Sensitivity-Invariance Visual Alignment framework that replaces operator-conditioned regularization with sample-wise, outcome-conditioned supervision. SIVA-RL constructs localized interventions through token-aligned, distance-constrained within-image PatchSwap. A frozen audit policy then scores each clean-intervention pair, and the observed reward drop becomes soft routing weights. Large-drop pairs drive sensitivity alignment, low-drop pairs drive clean-anchored invariance alignment, and ambiguous pairs are down-weighted. This design decouples intervention construction from supervision assignment and is compatible with both GRPO and DAPO backbones. Across nine multimodal reasoning benchmarks spanning mathematical, logical, and vision-dependent tasks, SIVA-RL improves 3B and 7B models over matched RL baselines in every setting. It yields an 8.79 percentage-point gain on vision-dependent reasoning and up to 14.9% relative overall improvement across all four GRPO- and DAPO-based configurations.
Chinese Translation
具有可验证奖励的强化学习（RLVR）推动了多模态推理，但答案级别的正确性并不能保证视觉-语言模型将其预测基于视觉证据。现有的视觉干预方法对比了原始图像和修改图像上的策略行为，但根据干预类型而非其观察到的效果进行监督。这一假设是错误的：相同的操作符在不同样本中产生异质的结果。我们提出了SIVA-RL，一个敏感性不变视觉对齐框架，它用样本级、结果条件的监督替代了操作符条件的正则化。SIVA-RL通过令牌对齐、距离约束的图像内PatchSwap构建局部干预。然后，一个冻结的审计策略对每对干净干预进行评分，观察到的奖励下降成为软路由权重。大幅下降的对驱动敏感性对齐，低下降的对驱动干净锚定的不变性对齐，而模糊的对则被降低权重。该设计将干预构建与监督分配解耦，并与GRPO和DAPO骨干网兼容。在涵盖数学、逻辑和依赖视觉的任务的九个多模态推理基准中，SIVA-RL在每个设置中都提高了3B和7B模型相较于匹配的RL基线。它在依赖视觉的推理上获得了8.79个百分点的提升，并在所有四种基于GRPO和DAPO的配置中实现了高达14.9%的相对整体改善。
cs.CV / 76 / 2607.13936
A novel unsupervised machine learning strategy to handle multimodal cardiac PET/MRI data
一种新颖的无监督机器学习策略用于处理多模态心脏PET/MRI数据
Abstract
Arrhythmogenic left ventricular cardiomyopathy is a genetic myocardial disease difficult to diagnose due to the lack of gold standard criteria. Simultaneous PET/MR imaging, combined with multiparametric quantitative analysis, could facilitate the identification of different profiles related to the phenotype and progression of cardiomyopathy. This preliminary study focuses on a methodological strategy for dealing with PET/MRI data, including inter-patient data linkage and regional analysis. Two-step clustering was applied to T1 and T2 maps, LGE, and 18F-FDG-PET images of 99 patients genetically diagnosed with arrhythmogenic left ventricular cardiomyopathy. Each patient's images were independently z-scored and summed into a single volume, which was clustered into supervoxels. Thirty-two inter-patient groups of supervoxels were obtained by spectral clustering. An "abnormality" score was assigned to each cluster and modality, and used to visualise abnormal regions likely associated with disease. They enabled the generation of automated textual and bullseye health reports for each patient, which were compared with cardiac imager assessments using balanced accuracy in repeated nested cross-validation. This approach was further validated on a larger cohort of 167 numerical phantoms. The reports generated by clustering accurately identified most of the cardiac physicians' observations (BA = 0.76 $\pm$ 0.04 in repeated nested cross-validation on patients, and BA $\ge$ 0.8 on phantoms). Furthermore, the identified abnormal clusters closely matched their visual observations, facilitating the identification of varying degrees of fibrosis or inflammation on the images. This approach enables a more systematic handling of multimodal PET/MRI data to characterise myocardial heterogeneity in arrhythmogenic left ventricular cardiomyopathy patients.
Chinese Translation
致心律失常的左心室心肌病是一种遗传性心肌疾病，由于缺乏金标准标准，诊断较为困难。同步的PET/MR成像结合多参数定量分析，可以促进识别与心肌病表型及其进展相关的不同特征。本初步研究聚焦于处理PET/MRI数据的方法策略，包括患者间数据关联和区域分析。对99名经基因诊断为致心律失常的左心室心肌病患者的T1和T2图谱、LGE及18F-FDG-PET图像应用了两步聚类。每位患者的图像独立进行z-score标准化并汇总为一个单一体积，然后聚类为超体素。通过谱聚类获得了32个患者间超体素组。为每个聚类和模态分配了“异常”评分，并用于可视化可能与疾病相关的异常区域。这些评分生成了每位患者的自动化文本和靶心健康报告，并与心脏影像学评估进行了比较，使用平衡准确率在重复嵌套交叉验证中进行评估。该方法在167个数值幻影的更大队列中进一步验证。聚类生成的报告准确识别了大多数心脏医生的观察结果（在患者的重复嵌套交叉验证中BA = 0.76 ± 0.04，幻影中BA ≥ 0.8）。此外，识别的异常聚类与其视觉观察高度一致，促进了对图像中不同程度纤维化或炎症的识别。这种方法使得对致心律失常的左心室心肌病患者的多模态PET/MRI数据进行更系统的处理成为可能，从而表征心肌的异质性。
cs.CV / 77 / 2607.13941
Peak-End-Net: A Peak-End Rule Inspired Framework for Generalizable Video Aesthetic Assessment
Peak-End-Net：一种受峰值-结束法则启发的可推广视频美学评估框架
Abstract
Video aesthetic assessment (VAA) aims to predict how aesthetically pleasing a video is, yet remains far less explored than other visual assessment tasks. Its progress is hindered not only by the scarcity of large-scale benchmarks, but also by the intrinsic subjectivity of aesthetic judgment, which is shaped by human perception. In this paper, we revisit VAA from a psychological perspective and propose \textit{Peak-End-Net}, a lightweight and interpretable framework inspired by the \textit{peak-end rule}, which suggests that people tend to judge a temporal experience mainly according to its salient moments and the ending. Building on this intuition, we first transfer knowledge from image aesthetic assessment (IAA) to VAA by introducing a pretrained IAA head to produce frame-wise aesthetic priors, which serve as surrogate signals for identifying aesthetically salient moments and guiding \textit{peak-end rule}-based temporal aggregation. To further capture how a video evolves aesthetically over time, we design an aesthetic rhythm encoder that models temporal progression beyond isolated moments. Additionally, we refine the overall assessment through a dynamic gated fusion mechanism to improve robustness under distribution shift. Our method is built on a frozen vision transformer (ViT) and requires only a small number of trainable parameters, making it scalable and parameter-efficient. Extensive experiments on two existing VAA benchmarks, including in-domain evaluation on VADB and cross-domain testing on DIVIDE-3K, demonstrate that our approach achieves state-of-the-art performance, affirming the value of psychologically grounded modeling for VAA. Our code and models are available at https://github.com/AMAP-ML/Peak-End-Net.
Chinese Translation
视频美学评估（VAA）旨在预测视频的美学愉悦度，但相较于其他视觉评估任务，其研究仍然较为不足。其进展不仅受到大规模基准稀缺的制约，还受到美学判断固有的主观性的影响，这种主观性受到人类感知的塑造。在本文中，我们从心理学的角度重新审视VAA，并提出了 extit{Peak-End-Net}，这是一个轻量且可解释的框架，受 extit{峰值-结束法则}的启发，该法则表明人们倾向于主要根据经历中的显著时刻和结尾来判断时间体验。基于这一直觉，我们首先通过引入一个预训练的图像美学评估（IAA）头，将知识从图像美学评估转移到视频美学评估，以生成逐帧的美学先验，这些先验作为识别美学显著时刻的替代信号，并指导基于 extit{峰值-结束法则}的时间聚合。为了进一步捕捉视频在时间上的美学演变，我们设计了一个美学节奏编码器，建模超越孤立时刻的时间进程。此外，我们通过动态门控融合机制优化整体评估，以提高在分布变化下的鲁棒性。我们的方法基于冻结的视觉变换器（ViT），仅需少量可训练参数，具有可扩展性和参数效率。在两个现有的VAA基准上进行的广泛实验，包括在VADB上的领域内评估和在DIVIDE-3K上的跨领域测试，证明了我们的方法达到了最先进的性能，确认了基于心理学建模在VAA中的价值。我们的代码和模型可在https://github.com/AMAP-ML/Peak-End-Net获取。
cs.CV / 78 / 2607.13945
PlumeQuant: Uncertainty-aware consistency assessment of methane plume masks and emission-rate estimates
PlumeQuant：基于不确定性的甲烷羽流掩膜和排放率估计的一致性评估
Abstract
Imaging spectrometers increasingly distribute source-resolved methane plume products in which the plume mask, integrated mass enhancement (IME), plume length, emission rate, and uncertainty are physically and algorithmically linked. Using 63 EMIT-derived Carbon Mapper plume records from 27 scenes, we show that these published scalar quantities do not uniquely constrain the plume boundary: substantially different yet plausible masks reproduce the same IME, plume length, and emission rate. Genetic-algorithm (GA) ensembles conditioned on the published IME and plume length make this equifinality explicit: the high-confidence core selected by nearly all target-consistent masks covers a median of 13% of the plausible footprint envelope, and ambiguity is largest for weak, low-overlap plumes. The diagnostics come from PlumeQuant, which recomputes IME, plume length, emission rate, and five-term uncertainty from distributed product components under stated conventions and evaluates four mask representations: the distributed reference mask, a transparent Carbon Mapper-informed analogue (CM-like), the GA ensemble, and optional expert edits. The CM-like mask is generated per plume without access to the reference mask or published quantities, with settings fixed once on a scene-disjoint 44-plume development split. It reproduced published IME with +0.72% median difference and emission rate with +0.16% (6.98% mean absolute), reached 0.843 median intersection-over-union against the reference masks, and matched the published uncertainty scale (median ratio 1.01). Holdout mean absolute errors were 7.6% (IME), 9.5% (length), and 6.1% (rate). These are product-level consistency diagnostics, not independent validation. They flag weak, offset, or ambiguous plumes for expert review.
Chinese Translation
成像光谱仪越来越多地分发源解析的甲烷羽流产品，其中羽流掩膜、集成质量增强（IME）、羽流长度、排放率和不确定性在物理和算法上是相互关联的。通过使用来自27个场景的63个EMIT衍生的碳映射器羽流记录，我们展示了这些已发布的标量量并不能唯一约束羽流边界：显著不同但合理的掩膜可以重现相同的IME、羽流长度和排放率。基于已发布的IME和羽流长度的遗传算法（GA）集成明确了这种等效性：几乎所有目标一致掩膜选择的高置信核心覆盖了合理足迹包络的中位数13%，而对于弱、低重叠的羽流，模糊性最大。这些诊断来自PlumeQuant，该工具根据既定约定重新计算IME、羽流长度、排放率和五项不确定性，并评估四种掩膜表示：分布式参考掩膜、透明的碳映射器信息类比（CM-like）、GA集成和可选的专家编辑。CM-like掩膜在没有访问参考掩膜或已发布量的情况下为每个羽流生成，设置在一个场景不重叠的44个羽流开发分割上固定。它重现了已发布的IME，具有+0.72%的中位差异和排放率+0.16%（6.98%的平均绝对差异），与参考掩膜的中位交并比达到0.843，并匹配了已发布的不确定性尺度（中位比率1.01）。保留的平均绝对误差为7.6%（IME）、9.5%（长度）和6.1%（排放率）。这些是产品级一致性诊断，而不是独立验证。它们标记出弱、偏移或模糊的羽流以供专家审查。
cs.CV / 79 / 2607.13976
CF-Net: Conflict Fusion with Speaker Normalisation and Certainty Weighting for Ambivalence/Hesitancy Recognition
CF-Net：通过说话者规范化和确定性加权进行冲突融合以识别矛盾/犹豫
Abstract
Detecting ambivalence and hesitancy (AH) in unconstrained video is challenging because the target signal is inherently ambiguous and expressed through subtle cross-modal incongruence rather than prototypical affect. We present CF-Net, a deep multimodal network submitted to the 3rd Edition of the AH Video Recognition Challenge (ABAW 11th, ECCV 2026), targeting the BAH dataset. CF-Net encodes visual, audio, and transcript streams with frozen SigLIP2, HuBERT, and DistilBERT backbones, normalises backbone features per speaker to reduce identity leakage, and fuses them via a ConflictFusion module that explicitly computes pairwise cross-modal incongruence. Training combines certainty-weighted focal loss, manifold mixup, and modality dropout; an auxiliary certainty-regression head leverages ambiguity annotations to stabilise learning on genuinely borderline samples. CF-Net achieves a Macro F1 of 0.7155 on the BAH validation set and 0.7364 (AP = 0.7492) on the private challenge test set.
Chinese Translation
在非约束视频中检测矛盾和犹豫（AH）是具有挑战性的，因为目标信号本质上是模糊的，并通过微妙的跨模态不一致而非典型的情感来表达。我们提出了CF-Net，这是一种深度多模态网络，提交至第3届AH视频识别挑战赛（ABAW第11届，ECCV 2026），针对BAH数据集。CF-Net使用冻结的SigLIP2、HuBERT和DistilBERT骨干网络对视觉、音频和文本流进行编码，按说话者规范化骨干特征以减少身份泄漏，并通过冲突融合模块（ConflictFusion）融合这些特征，该模块显式计算成对的跨模态不一致。训练结合了确定性加权的焦点损失、流形混合和模态丢弃；一个辅助的确定性回归头利用模糊注释来稳定在真正边缘样本上的学习。CF-Net在BAH验证集上实现了0.7155的宏F1分数，在私有挑战测试集上达到了0.7364（AP = 0.7492）。
cs.CV / 80 / 2607.13978
Music-to-Dance Generation via Atomic Movements
基于原子动作的音乐舞蹈生成
Abstract
Music-driven dance generation aims to produce human motion that is both rhythmically synchronized and semantically consistent with music. While recent neural approaches have achieved impressive visual realism, they typically model motion as a continuous signal and neglect its compositional nature, making generated dances structurally incoherent and difficult to control. In this work, we introduce a structure-aware framework that models choreography as a sequence of atomic movements-semantically interpretable motion events that serve as the building blocks of dance. To construct this atomic movement vocabulary, we first segment large-scale dance data and cluster them into atomic movement groups. We then employ a large language model to semantically relabel and refine the clusters, yielding a set of interpretable and reusable atomic movements. Based on these atomic movement annotations, we design a two-stage generation framework that mirrors the human choreography process. In the atomic movement planning stage, the model predicts the type, duration, and timing of atomic movements conditioned on the input music, forming a symbolic dance allocation. In the completion stage, a transition-aware generator synthesizes smooth and stylistically coherent motion conditioned on the planned structure. Extensive experiments demonstrate that our method produces dances with significantly improved structural coherence, rhythmic alignment, and perceptual naturalness compared to existing baselines, while providing enhanced interpretability and controllable editing through explicit structural representation.
Chinese Translation
音乐驱动的舞蹈生成旨在产生与音乐在节奏上同步且在语义上一致的人类动作。尽管最近的神经网络方法在视觉真实感方面取得了令人印象深刻的成果，但它们通常将动作建模为连续信号，忽视了其组合特性，从而导致生成的舞蹈结构不连贯且难以控制。在本研究中，我们提出了一种结构感知框架，将编舞建模为一系列原子动作——可语义解释的运动事件，作为舞蹈的基本构建块。为了构建这一原子动作词汇，我们首先对大规模舞蹈数据进行分段，并将其聚类为原子动作组。然后，我们利用大型语言模型对聚类进行语义重新标注和精炼，从而生成一组可解释且可重复使用的原子动作。基于这些原子动作注释，我们设计了一个两阶段生成框架，模拟人类编舞过程。在原子动作规划阶段，模型根据输入音乐预测原子动作的类型、持续时间和时机，形成符号舞蹈分配。在完成阶段，过渡感知生成器根据规划的结构合成平滑且风格一致的运动。大量实验表明，与现有基准相比，我们的方法生成的舞蹈在结构连贯性、节奏对齐和感知自然性方面显著提升，同时通过明确的结构表示提供了更好的可解释性和可控编辑。
cs.CV / 81 / 2607.13983
Screening Is Effective for Visual Recognition
筛选在视觉识别中的有效性
Abstract
Vision Transformer (ViT) has been widely used as a powerful framework for modeling global dependencies among image patches. However, its core component, self-attention assigns softmax-normalized relative weights to all patches, making it difficult to evaluate the relevance between patches independently. In visual recognition, images often contain many background or redundant patches, yet self-attention cannot explicitly reject such irrelevant patches, which may introduce unnecessary information into feature aggregation. To address this limitation, Screening has been proposed in the field of language modeling, where the relevance of each token is independently evaluated based on query-key similarity and low-relevance tokens are explicitly excluded through thresholding. In this work, we propose VisionScreen, a new vision model that extends Screening mechanism to visual recognition. VisionScreen treats image patches as tokens arranged on a two-dimensional grid and extends absolute relevance estimation based on query-key similarity to the two-dimensional spatial domain. This allows each patch to selectively aggregate only content-wise and spatially relevant patches without relying on competition among patches. Experiments on image classification benchmarks demonstrate that the proposed method outperforms conventional ViT. These results suggest that Screening can be effective for visual recognition, offering an alternative to relative feature aggregation based on softmax attention.
Chinese Translation
视觉变换器（Vision Transformer, ViT）作为一种强大的框架，已被广泛应用于建模图像块之间的全局依赖关系。然而，其核心组件自注意力机制对所有图像块分配了经过softmax归一化的相对权重，这使得独立评估图像块之间的相关性变得困难。在视觉识别中，图像通常包含许多背景或冗余的图像块，而自注意力机制无法明确排除这些无关的图像块，这可能会在特征聚合中引入不必要的信息。为了解决这一局限性，筛选（Screening）在语言建模领域被提出，其中每个标记的相关性是基于查询-键相似性独立评估的，低相关性的标记通过阈值明确排除。在本研究中，我们提出了VisionScreen，这是一种新的视觉模型，将筛选机制扩展到视觉识别中。VisionScreen将图像块视为在二维网格上排列的标记，并将基于查询-键相似性的绝对相关性估计扩展到二维空间域。这使得每个图像块能够选择性地聚合仅与内容和空间相关的图像块，而无需依赖于图像块之间的竞争。在图像分类基准测试中的实验表明，所提出的方法优于传统的ViT。这些结果表明，筛选在视觉识别中是有效的，为基于softmax注意力的相对特征聚合提供了一种替代方案。
cs.CV / 82 / 2607.13986
Task-Specific Feature Fusion Method for Multi-Task Affective Behavior Analysis
针对多任务情感行为分析的任务特定特征融合方法
Abstract
The 11th Affective Behavior Analysis in-the-wild (ABAW11) Multi-Task Learning Challenge requires a unified system to predict valence-arousal, categorical expressions, and facial action units from the official s-Aff-Wild2 images. Although these tasks are naturally related through facial behavior, our validation experiments show that they benefit from different visual features, temporal processing strategies, fusion mechanisms, and calibration procedures. In this paper, we study task-adaptive feature fusion for ABAW11 multi-task affective behavior analysis. We first adapt two pretrained visual backbones, DINOv2 ViT-L and DINOv3 ConvNeXt-base, on an external expression-oriented facial image set and then freeze them to extract complementary frame-level features from the official ABAW11 data. On top of these frozen features, we systematically compare frame-level prediction heads, temporal convolutional heads, post-hoc temporal smoothing, LightGBM models, feature concatenation, gated fusion, residual fusion, late logit fusion, threshold calibration, and shared MTL structures. The final system selects task-specific fusion and prediction strategies rather than forcing all tasks to share a single architecture. On the ABAW11 validation set, the selected system achieves an EXPR macro-F1 of 0.4222, an AU macro-F1 of 0.5402, and a mean VA CCC of 0.6717, resulting in an overall validation score of 1.6341. The results suggest that task-adaptive fusion of frozen visual features is a simple and effective strategy for ABAW-style multi-task affective behavior analysis.
Chinese Translation
第11届野外情感行为分析（ABAW11）多任务学习挑战要求一个统一的系统，从官方的s-Aff-Wild2图像中预测情感的愉悦度-唤醒度、类别表达和面部动作单元。尽管这些任务通过面部行为自然相关，但我们的验证实验表明，它们受益于不同的视觉特征、时间处理策略、融合机制和校准程序。本文研究了针对ABAW11多任务情感行为分析的任务自适应特征融合。我们首先在一个外部以表情为导向的面部图像集上调整两个预训练的视觉主干网络，DINOv2 ViT-L和DINOv3 ConvNeXt-base，然后将其冻结，以从官方的ABAW11数据中提取互补的帧级特征。在这些冻结特征的基础上，我们系统地比较了帧级预测头、时间卷积头、后处理时间平滑、LightGBM模型、特征拼接、门控融合、残差融合、后期对数融合、阈值校准和共享多任务学习结构。最终系统选择了任务特定的融合和预测策略，而不是强迫所有任务共享单一架构。在ABAW11验证集上，所选系统实现了0.4222的EXPR宏F1、0.5402的AU宏F1和0.6717的平均VA CCC，最终验证得分为1.6341。结果表明，冻结视觉特征的任务自适应融合是一种简单有效的策略，适用于ABAW风格的多任务情感行为分析。
cs.CV / 83 / 2607.14005
M$^\text{4}$World: A Multi-view Multimodal Driving World Model for Interactive Object Manipulation and Minute-long Streaming
M$^ ext{4}$World：一种用于交互式物体操控和分钟级流媒体的多视角多模态驾驶世界模型
Abstract
Driving-world generation has emerged as a core capability for scalable autonomous-driving simulation, yet existing methods remain limited in object-level controllability and long-horizon stability. We present M$^\text{4}$World, a Multi-view and Multimodal generative driving world model that synthesizes future surround-view video streams and synchronized LiDAR scans while supporting interactive object Manipulation and stable Minute-long streaming. Fine-grained object manipulation is realized through a flexible conditioning interface that supports explicit control over both the spatial layout and visual appearance of individual objects. Stable minute-long streaming, on the other hand, is achieved through a multi-stage training framework that enables online causal generation in only four denoising steps while maintaining coherent world dynamics throughout extended rollouts. Building on these components, we introduce an efficient few-clip post-training as well as a suite of visual reference-conditioned generation models, preserving general generation ability while allowing rare-case customization for long-tail controllability. To assess controllability beyond realism, we further introduce an automated VLM-based judging pipeline that evaluates scene-level condition adherence, view-wise object controllability, and cross-view object consistency. Comprehensive experiments show that M$^\text{4}$World consistently delivers high generation quality, precise controllability, and stable minute-long streaming. Together with downstream long-tail augmentation and scene editing, these results demonstrate the potential of M$^\text{4}$World for controllable, scalable driving simulation.
Chinese Translation
驾驶世界生成已成为可扩展自主驾驶仿真的核心能力，然而现有方法在物体级可控性和长时间稳定性方面仍然有限。我们提出了M$^ ext{4}$World，这是一种多视角和多模态生成驾驶世界模型，能够合成未来的全景视频流和同步的激光雷达（LiDAR）扫描，同时支持交互式物体操控和稳定的分钟级流媒体。通过灵活的条件接口实现细粒度的物体操控，该接口支持对单个物体的空间布局和视觉外观进行显式控制。另一方面，稳定的分钟级流媒体通过多阶段训练框架实现，该框架仅需四个去噪步骤即可实现在线因果生成，同时在扩展的回放过程中保持一致的世界动态。在这些组件的基础上，我们引入了一种高效的少量剪辑后训练方法，以及一套视觉参考条件生成模型，既保留了通用生成能力，又允许对长尾可控性进行稀有案例的定制。为了评估超越现实主义的可控性，我们进一步引入了一种基于自动化视觉语言模型（VLM）的评判管道，该管道评估场景级条件遵循、视角级物体可控性和跨视角物体一致性。综合实验表明，M$^ ext{4}$World始终提供高质量的生成、精确的可控性和稳定的分钟级流媒体。结合下游长尾增强和场景编辑，这些结果展示了M$^ ext{4}$World在可控、可扩展驾驶仿真中的潜力。
cs.CV / 84 / 2607.14041
Multi-Expert Routing for Multi-Domain Low-Resource OCR: A Manchu Case Study
多专家路由用于多领域低资源OCR：满族案例研究
Abstract
Historical Manchu OCR must accommodate various visually distinct writing styles, including regular script, running script, and the semi-cursive chancery hand used in palace memorials, despite limited labeled data. We study a multi-expert system that reuses checkpoints from an iterative fine-tuning process as domain specialists and uses a lightweight page-level image classifier to dispatch pages by visual style. When the checkpoint pool lacks a suitable specialist, we train an additional expert for that domain. On three frozen test sets, the routed system matches the selected specialist for each style at two-decimal precision: 0.30 percent CER on regular script, 1.57 percent on memorials, and 4.83 percent on running script. The router achieves 99.3 percent page-level domain accuracy and matches the domain-label oracle at the same precision. Two of the three selected specialists were not trained specifically for their final domain; only the running-script expert was trained with that domain as its target. We report the evaluation protocol, router design, and per-page predictions to make the comparison reproducible.
Chinese Translation
历史满族OCR必须适应多种视觉上明显不同的书写风格，包括楷书、行书以及用于宫廷奏折的半行草书，尽管标注数据有限。我们研究了一种多专家系统，该系统重用来自迭代微调过程的检查点作为领域专家，并使用轻量级页面级图像分类器根据视觉风格分配页面。当检查点池缺乏合适的专家时，我们为该领域训练额外的专家。在三个冻结的测试集上，路由系统以两位小数的精度匹配每种风格的选定专家：楷书的字符错误率（CER）为0.30%，奏折为1.57%，行书为4.83%。路由器实现了99.3%的页面级领域准确率，并以相同的精度匹配领域标签的oracle。在三位选定专家中，有两位并未专门为其最终领域进行训练；只有行书专家是以该领域为目标进行训练的。我们报告了评估协议、路由器设计以及每页预测，以使比较具有可重复性。
cs.CV / 85 / 2607.14076
From Pixels to States: Rethinking Interactive World Models as Game Engines
从像素到状态：重新思考互动世界模型作为游戏引擎
Abstract
Building interactive worlds that respond coherently to player actions has long been a shared goal of computer graphics, games, and artificial intelligence. Recent video generative models provide a data-driven route toward this goal by predicting future observations conditioned on user actions, and are increasingly regarded as potential next-generation game engines. Realizing a genuinely interactive game world, however, requires interaction outcomes that follow rules over evolving game conditions, consequences that persist over long horizons, and a generation loop that operates in real time. Conventional game engines realize these properties through a recurrent action-state-observation loop, in which player actions update an explicit game state according to predefined rules and observations are rendered from the resulting state. Taking this loop as an organizing lens, this paper examines interactive game world modeling along four dimensions: player action control, game state dynamics, state-observation persistence, and real-time interactive generation. For each dimension, we start from the capabilities required by an interactive game world, group existing approaches into representative families, and discuss the strengths and trade-offs of each family. Complementing this analysis, we present a scalable data engine for Black Myth: Wukong that collects over 90 hours of gameplay with frame-aligned player actions, ground-truth game states, and visual observations, together with structured and semantic annotations, as a resource for state-aware game world modeling. We hope this paper offers a clear picture of where the field stands and fosters progress toward interactive game worlds.
Chinese Translation
构建能够对玩家行为做出连贯响应的互动世界，一直以来都是计算机图形学、游戏和人工智能的共同目标。近期的视频生成模型通过基于用户行为预测未来观察，提供了一条数据驱动的途径来实现这一目标，并越来越被视为潜在的下一代游戏引擎。然而，实现一个真正互动的游戏世界需要交互结果遵循不断变化的游戏条件下的规则，后果在较长时间内持续存在，以及一个实时运作的生成循环。传统游戏引擎通过一个递归的动作-状态-观察循环实现这些特性，其中玩家的行为根据预定义规则更新明确的游戏状态，并从生成的状态中渲染观察。本文以此循环为组织视角，从四个维度考察互动游戏世界建模：玩家行为控制、游戏状态动态、状态-观察持久性和实时互动生成。对于每个维度，我们从互动游戏世界所需的能力出发，将现有方法分组为代表性家族，并讨论每个家族的优缺点。作为对这一分析的补充，我们为《黑神话：悟空》提供了一个可扩展的数据引擎，收集了超过90小时的游戏玩法，包含帧对齐的玩家行为、真实游戏状态和视觉观察，以及结构化和语义注释，作为状态感知游戏世界建模的资源。我们希望本文能够清晰展示该领域的现状，并促进互动游戏世界的发展。
cs.CV / 86 / 2607.14088
VideoRAE: Taming Video Foundation Models for Generative Modeling via Representation Autoencoders
VideoRAE：通过表示自编码器驯化视频基础模型以进行生成建模
Abstract
Video generative models commonly rely on latent spaces learned by 3D Variational Autoencoders (3D-VAEs). However, conventional 3D-VAEs are mainly optimized for pixel-level reconstruction, which can limit the semantic and spatio-temporal structure captured by their latents. Meanwhile, Video Foundation Models (VFMs) such as V-JEPA 2 and VideoMAEv2 show strong video understanding capabilities, yet whether their frozen representations can be transformed into compact, reconstruction-capable, and generation-friendly video latents remains largely unexplored. We answer this question with VideoRAE, a representation autoencoder that leverages multi-scale hierarchical features from a frozen video foundation encoder and compresses them with a lightweight 1D self-attention projector. VideoRAE supports both continuous latents for Diffusion Transformers and discrete tokens for autoregressive models via multi-codebook high-dimensional quantization. During decoding, a local-and-global representation alignment objective with the frozen VFM teacher improves semantic preservation and enables training without KL regularization. Experiments show that VideoRAE achieves strong reconstruction in both continuous and discrete regimes. On UCF-101, it obtains state-of-the-art class-to-video gFVDs of 40 and 93 with AR and DiT generators, respectively, while converging approximately 5x faster than competing autoencoder baselines. In a controlled 2B-scale text-to-video study, replacing LTX-VAE with VideoRAE leads to faster convergence under comparable settings. These results validate frozen VFM representations as versatile and generation-friendly video latents. The model and code will be released on https://zhxie0117.github.io/VideoRAE.
Chinese Translation
视频生成模型通常依赖于由三维变分自编码器（3D-VAEs）学习的潜在空间。然而，传统的3D-VAEs主要针对像素级重建进行优化，这可能限制其潜在空间所捕捉的语义和时空结构。同时，视频基础模型（VFMs）如 V-JEPA 2 和 VideoMAEv2 展现出强大的视频理解能力，但其冻结表示是否能够转化为紧凑的、具备重建能力和生成友好的视频潜在空间仍然未被充分探索。我们通过 VideoRAE 来回答这个问题，VideoRAE 是一种表示自编码器，利用来自冻结视频基础编码器的多尺度层次特征，并通过轻量级的一维自注意力投影器进行压缩。VideoRAE 支持用于扩散变换器的连续潜在空间和用于自回归模型的离散标记，通过多代码本高维量化实现。在解码过程中，结合冻结 VFM 教师的局部与全局表示对齐目标，改善了语义保留，并使得在没有 KL 正则化的情况下进行训练成为可能。实验表明，VideoRAE 在连续和离散模式下均实现了强大的重建能力。在 UCF-101 数据集上，它分别以 AR 和 DiT 生成器获得了 40 和 93 的最先进类到视频 gFVD，同时收敛速度约为竞争性自编码器基线的 5 倍。在一个受控的 2B 规模文本到视频研究中，用 VideoRAE 替换 LTX-VAE 在可比设置下实现了更快的收敛。这些结果验证了冻结 VFM 表示作为多功能且生成友好的视频潜在空间的有效性。模型和代码将发布在 https://zhxie0117.github.io/VideoRAE。
cs.AI / 1 / 2607.13037
OriginBlame: Record- and Token-Level Data Provenance for AI Training Datasets
OriginBlame：用于人工智能训练数据集的记录级和标记级数据来源
Abstract
When a data contributor requests removal, model trainers face a practical gap: unlearning algorithms require a forget set, yet no tool can locate which training records belong to a given author. Existing provenance systems operate at file or dataset level, forcing catastrophic over-deletion. We present ob, a record- and token-level data provenance system that propagates author identity through data processing pipelines and resolves revocation requests into precise forget sets via deterministic queries. Evaluation on 219,555 Wikipedia pages demonstrates that record-level provenance eliminates dataset-level over-deletion (from 101x to 1.3x), while integration adds 1.3-4.0% throughput overhead (HuggingFace) and 2.1-19.0% (Datatrove) on wiki data. On a 1.7B model, provenance-based forget sets improve unlearning by 42% over random baselines.
Chinese Translation
当数据贡献者请求删除时，模型训练者面临一个实际的难题：去学习算法需要一个遗忘集，但没有工具能够定位哪些训练记录属于特定作者。现有的数据来源系统仅在文件或数据集层面操作，导致灾难性的过度删除。我们提出了 ob，一个记录级和标记级的数据来源系统，它通过数据处理管道传播作者身份，并通过确定性查询将撤销请求解析为精确的遗忘集。在对 219,555 个维基百科页面的评估中，记录级数据来源消除了数据集级的过度删除（从 101 倍降低到 1.3 倍），而集成增加了 1.3-4.0% 的吞吐量开销（HuggingFace）和 2.1-19.0%（Datatrove）在维基数据上。在一个 17 亿参数的模型上，基于数据来源的遗忘集比随机基线提高了 42% 的去学习效果。
cs.AI / 2 / 2607.13049
SPINE: Bridging the Cyber-Physical Gap with Agentic AI
SPINE：通过自主智能弥合网络物理差距
Abstract
Foundation models have given robots a sophisticated brain for complex decision-making, yet deploying that intelligence into a physical platform still demands tedious, expert-driven calibration. This deployment gap, the robot's spinal cord, remains a primary bottleneck to scalable Embodied AI. Hence, we propose SPINE (Scalable Physical Integration with ageNtic Expertise): an agentic framework for systematically debugging and deploying bimanual robots with minimal robotics expertise. SPINE's harness comprises two orchestrated multi-agent workflows: a profile builder that creates robot-specific context, and a debugger that cycles through diagnosis, repair, and validation until teleoperation works. Across seven DOBOT X-Trainer debugging scenarios, a robotics novice using SPINE outperformed human operators using Claude Code with the same reference materials, but without SPINE's structured workflow, improving operationalization success from 75% to 100% and reducing mean time-to-teleoperation from 16 min 45 s to 13 min 47 s. On AgileX PiPER, a distinct ROS/CAN bimanual arm, SPINE resolved all 10 implanted bugs, versus 9 out of 10 for the expert baseline, in nearly the same amount of time. Together, these results show that SPINE can transfer across bimanual platforms, reduce dependence on expert calibration, and move embodied AI closer to scalable real-world deployment.
Chinese Translation
基础模型为机器人提供了复杂决策所需的高级智能，但将这种智能部署到物理平台仍然需要繁琐的专家驱动校准。这一部署差距，即机器人的脊髓，仍然是可扩展具身人工智能的主要瓶颈。因此，我们提出了SPINE（可扩展物理集成与自主专业知识）：一个自主框架，用于系统性调试和部署双手机器人，所需的机器人专业知识最少。SPINE的框架包括两个协调的多智能体工作流程：一个创建机器人特定上下文的配置生成器，以及一个在诊断、修复和验证之间循环，直到远程操作成功的调试器。在七个DOBOT X-Trainer调试场景中，使用SPINE的机器人新手在相同参考材料下超越了使用Claude Code的人类操作员，但没有SPINE的结构化工作流程，将操作成功率从75%提高到100%，并将平均远程操作时间从16分钟45秒减少到13分钟47秒。在AgileX PiPER，一个独特的ROS/CAN双手臂上，SPINE解决了所有10个植入的错误，而专家基线仅解决了9个，所用时间几乎相同。这些结果表明，SPINE可以在双手平台之间转移，减少对专家校准的依赖，并使具身人工智能更接近可扩展的现实世界部署。
cs.AI / 3 / 2607.13069
Interventional Grounding Audits: Black-Box Premise-Dependency Tests for LLM Chain-of-Thought via Predicate Substitution
干预性基础审计：通过谓词替换对大型语言模型链式思维进行黑箱前提依赖性测试
Abstract
Large language models produce chain-of-thought (CoT) reasoning that appears logically sound yet may not genuinely depend on its stated premises. We introduce interventional grounding audits, a black-box, step-level test of premise dependency: we intervene on a single premise by substituting its target predicate with a fresh symbol, re-run the model, and check whether each reasoning step's normalized conclusion (canonical predicate form) changes. We evaluate on ProntoQA, a synthetic multi-hop deductive reasoning benchmark with gold proof trees, where step-level premise dependencies are known. Applied to 50 ProntoQA problems with GPT-4o, our method achieves F1 = 0.806 on detecting proof-tree dependencies (F1 = 0.885 on predicate-determining dependencies; Recall = 100%), significantly outperforming a self-consistency baseline (F1 = 0.343; 95% bootstrap CIs non-overlapping). We further identify that 66% of correctly-solved problems contain at least one aligned step insensitive to a direct proof-tree dependency under consistent substitution -- all involving entity-introduction premises, a documented blind spot of the consistent-substitution evaluator -- a "right answer, wrong reasoning" signal invisible to passive methods. All audit certificates, raw outputs, and reproduction scripts are available in a public GitHub repository, and we discuss scope limits beyond formal, parsable benchmarks.
Chinese Translation
大型语言模型生成的链式思维（CoT）推理看似在逻辑上是合理的，但可能并不真正依赖于其所陈述的前提。我们引入了干预性基础审计，这是一种黑箱的逐步测试前提依赖性的方法：我们通过用新符号替换目标谓词来干预单个前提，重新运行模型，并检查每个推理步骤的归一化结论（标准谓词形式）是否发生变化。我们在 ProntoQA 上进行评估，这是一个具有金证据树的合成多跳演绎推理基准，其中已知逐步前提依赖性。应用于 50 个 ProntoQA 问题与 GPT-4o，我们的方法在检测证据树依赖性方面达到了 F1 = 0.806（在确定谓词依赖性方面 F1 = 0.885；召回率 = 100%），显著优于自一致性基线（F1 = 0.343；95% 自助法置信区间不重叠）。我们进一步发现，66% 的正确解决问题包含至少一个在一致替换下对直接证据树依赖性不敏感的对齐步骤——所有这些都涉及实体引入前提，这是一致替换评估器的一个已记录盲点——这是一个“正确答案，错误推理”的信号，无法通过被动方法察觉。所有审计证书、原始输出和重现脚本均可在公共 GitHub 仓库中获取，我们讨论了超越正式、可解析基准的范围限制。
cs.AI / 4 / 2607.13073
Probabilistic Extension of Neuro-Symbolic AGI Robots based on Belnap's Typed Intensional FOL
基于贝尔纳普类型意向性一阶逻辑的神经符号AGI机器人的概率扩展
Abstract
Neuro-symbolic AI based on $IFOL_B$ is a way to combine neural learning and symbolic reasoning to overcome limitations of purely neural systems (like lack of interpretability and logical structure) with formal logical machinery for self-reference. In this paper we expand the cognitive power of $IFOL_B$ by using the probability computation for the currently unknown sentences, based on Nilsson's probability structure for the $IFOL_B$. We introduce the global symmetry transformation that preserves the current knowledge database and logical deduction, and the local one used for real-time decisions about concrete (sub)problems that involve only a very strict subset of $IFOL_B$ predicates. The computation of probability density function $KI$ in both cases, based on the Shannon's maximum information entropy, is provided by neural networks of this probabilistic neuro-symbolic AGI.
Chinese Translation
基于 $IFOL_B$ 的神经符号人工智能是一种结合神经学习和符号推理的方法，旨在克服纯神经系统的局限性（如缺乏可解释性和逻辑结构）以及用于自我引用的形式逻辑机制。本文通过使用基于Nilsson概率结构的概率计算，扩展了 $IFOL_B$ 的认知能力，以处理当前未知的句子。我们引入了全局对称变换，以保持当前知识数据库和逻辑推理的完整性，以及用于实时决策的局部变换，这些决策涉及仅包含 $IFOL_B$ 谓词的一个非常严格的子集。基于香农最大信息熵的概率密度函数 $KI$ 的计算在这两种情况下均由该概率神经符号AGI的神经网络提供。
cs.AI / 5 / 2607.13104
Self-Improvements in Modern Agentic Systems: A Survey
现代自主系统中的自我改进：一项调查
Abstract
Self-improving autonomous agents are moving from research prototypes to deployed systems. The primary goal is controllable evolution, or adaptation, from experience with minimal or even no human input. This survey frames modern self-improving agents as adaptive systems that convert experience into accumulated capability gains. We offer a system-level framework that represents a modern agent as a configuration coupling a foundation model with an operational scaffold of prompts, memory, tools, and control logic. Within this framework, self-improvement is formalized as a self-induced update operator that obtains and commits updates to model parameters or scaffold components. We organize prior work by update target and by the signals that drive change, then review applications and discuss evaluation, before closing with open problems and future directions. For convenience, we track technical updates on https://github.com/selfimproving-agent/awesome-Self-Improving-Agents.
Chinese Translation
自我改进的自主代理正在从研究原型转向实际部署系统。其主要目标是可控的演变或适应，依靠经验实现最小甚至无人的输入。本调查将现代自我改进代理框定为将经验转化为累积能力提升的适应性系统。我们提供了一个系统级框架，将现代代理表示为一个配置，结合了基础模型与操作支架，包括提示、记忆、工具和控制逻辑。在该框架内，自我改进被形式化为一个自我诱导的更新算子，该算子获取并提交对模型参数或支架组件的更新。我们根据更新目标和驱动变化的信号组织了以往的工作，随后回顾应用并讨论评估，最后总结开放问题和未来方向。为方便起见，我们在 https://github.com/selfimproving-agent/awesome-Self-Improving-Agents 上跟踪技术更新。
cs.AI / 6 / 2607.13115
Improving Molecular Property Prediction in Small Language Models Using Graph-based Tools
利用基于图的工具提升小型语言模型中的分子性质预测
Abstract
Small language models (SLMs) have shown promise for zero-shot molecular property prediction from SMILES strings, yet they often suffer from structural blindness because sequence representations under-specify key graph-topological cues. We propose a modular Context-Augmented Prompting framework that enables agentic tool use at inference time: a trained GNN expert model provides a predictive hint with confidence, and a GNN extracts an instance-specific explanatory subgraph (e.g., a subgraph SMILES and an accompanying explanatory paragraph). We evaluate three commonly used SLMs on MUTAG and Tox21 under five prompting configurations ranging from SMILES-only to using all available tools at hand. Across two datasets, enriching prompts with graph-derived context yields substantial accuracy gains, often exceeding 25% relative improvement and up to 74% on Tox21. We further validate the functional relevance of the extracted motifs via a necessity-based edge-drop intervention. Despite the observed gains, a persistent gap remains to specialized GNN models, highlighting both the value and limits of text-conditioned reasoning for molecular structure.
Chinese Translation
小型语言模型（SLMs）在从SMILES字符串进行零-shot分子性质预测方面显示出潜力，但它们常常因结构盲目性而受到影响，因为序列表示不足以明确关键的图拓扑线索。我们提出了一种模块化的上下文增强提示框架，该框架在推理时能够实现自主工具使用：一个训练好的图神经网络（GNN）专家模型提供具有信心的预测提示，而GNN提取特定实例的解释性子图（例如，子图SMILES及其伴随的解释段落）。我们在MUTAG和Tox21上评估了三种常用的SLMs，采用从仅使用SMILES到使用所有可用工具的五种提示配置。在两个数据集上，使用图生成的上下文丰富提示显著提高了准确性，通常超过25%的相对提升，在Tox21上甚至达到74%。我们进一步通过基于必要性的边缘删除干预验证了提取的图案的功能相关性。尽管观察到了提升，但与专门的GNN模型之间仍存在持续的差距，突显了文本条件推理在分子结构中的价值和局限性。
cs.AI / 7 / 2607.13157
Oracle Agent Memory as an Enterprise Memory Substrate for Long-Horizon AI Agents
Oracle代理记忆作为长时程人工智能代理的企业记忆基础
Abstract
Agent memory is a systems problem for long-horizon agents. Practical deployments require retention of task state across extended conversations, recovery of user-specific facts and preferences across sessions, and accumulation of procedural knowledge from prior outcomes. These requirements extend beyond document retrieval: a memory layer must determine which interactions become durable state, how that state is scoped, how it is retrieved under latency constraints, and how it is revised or removed over time. This report studies Oracle Agent Memory as a database-native memory substrate built on Oracle Database. Three themes organize the discussion: memory as a lifecycle spanning ingestion, extraction, consolidation, retrieval, summarization, and revision or removal; a layered architecture that separates an active memory core from a passive memory-store interface with explicit scope control across users, agents, and threads; and evaluation methodology in which downstream task accuracy is complemented by memory-centric measures such as evidence retrieval, recall, latency, and estimated token use. The report summarizes LongMemEval results, reaching 93.8% accuracy, compares Oracle Agent Memory against flat-history baselines, using about 10.7x fewer tokens, and published or reported external baselines where available, and closes with implementation-oriented appendix material covering setup, thread lifecycle, and search semantics.
Chinese Translation
代理记忆是长时程代理的一个系统性问题。实际部署需要在延续的对话中保持任务状态，跨会话恢复用户特定的事实和偏好，以及从先前结果中积累程序性知识。这些要求超出了文档检索的范围：一个记忆层必须决定哪些交互成为持久状态，如何界定该状态，如何在延迟约束下进行检索，以及如何随着时间的推移进行修订或删除。本报告研究了Oracle代理记忆作为建立在Oracle数据库上的数据库原生记忆基础。讨论围绕三个主题展开：记忆作为一个生命周期，涵盖摄取、提取、整合、检索、总结以及修订或删除；一个分层架构，将主动记忆核心与被动记忆存储接口分开，并在用户、代理和线程之间进行明确的范围控制；以及评估方法论，其中下游任务的准确性与以记忆为中心的指标（如证据检索、召回率、延迟和估计的令牌使用）相辅相成。报告总结了LongMemEval的结果，达到了93.8%的准确率，并将Oracle代理记忆与平面历史基线进行了比较，使用的令牌数量约为10.7倍更少，并与已发布或报告的外部基线进行了对比，最后附录部分提供了与实施相关的材料，包括设置、线程生命周期和搜索语义。
cs.AI / 8 / 2607.13172
Learning Safe Agent Behaviour from Human Preferences and Justifications via World Models
从人类偏好和理由中学习安全代理行为通过世界模型
Abstract
We address the problem of safely training an agent policy and deploying a good and safe policy, in settings where the environment dynamics are unknown and no suitable reward function is available. In the context of safety-critical environments, we consider traditional reinforcement learning impractical and resort to the resource of human input. We introduce DROPJ, a human-centred method for both safe training and deployment. We first learn a world model (a learned simulator) from a dataset of prior real-world trajectories. A human then plays the game in this learned simulator to extract several informative simulated trajectories. From these, we sample pairs of simulated trajectory segments and elicit from a human their preference over these segments, as well as a reason (justification) for their choice. We then train a reward model from these justified preferences and use it, together with the world model, to directly deploy the agent using model predictive control. Running real-user experiments, we find that generating informative simulated trajectories from a user significantly reduces the computational cost during training compared to other strategies, and can also improve the performance during deployment. In the context of training within a learned simulator, we show that the use of preferences rather than other types of feedback substantially improves the performance during deployment. We further demonstrate that safety justifications accompanying preferences can significantly enhance safety or prioritise user-prescribed aspects of safety associated with them during deployment.
Chinese Translation
我们解决了在环境动态未知且没有合适奖励函数的情况下，安全训练代理策略和部署良好且安全策略的问题。在安全关键环境中，我们认为传统的强化学习不切实际，因此依赖于人类输入的资源。我们提出了DROPJ，这是一种以人为中心的方法，用于安全训练和部署。我们首先从先前真实世界轨迹的数据集中学习一个世界模型（一个学习的模拟器）。然后，人类在这个学习的模拟器中进行游戏，以提取若干个信息丰富的模拟轨迹。从中，我们抽取模拟轨迹段的配对，并从人类那里获取他们对这些段的偏好以及选择的理由（理由）。然后，我们根据这些有理由的偏好训练一个奖励模型，并将其与世界模型结合，直接使用模型预测控制来部署代理。通过进行真实用户实验，我们发现，从用户生成信息丰富的模拟轨迹显著降低了训练过程中的计算成本，相较于其他策略，且在部署过程中也能提高性能。在学习的模拟器中进行训练的背景下，我们展示了使用偏好而非其他类型反馈显著提高了部署过程中的性能。我们进一步证明，伴随偏好的安全理由可以显著增强安全性或优先考虑与之相关的用户规定的安全方面。
cs.AI / 9 / 2607.13219
CayleyR: Solving the TopSpin puzzle via cycle intersection
CayleyR：通过循环交集解决 TopSpin 拼图
Abstract
We present cayleyR, an R package for solving permutation puzzles by detecting cycle intersections in Cayley graphs. The core algorithm performs an iterative bidirectional search: from both the initial and target permutation states, random operation sequences generate cycles in the Cayley graph of the symmetric group Sn; their intersection yields a connecting path. When no direct intersection is found, a distance-guided bridge selection narrows the gap, and the process repeats. The package targets the TopSpin(n,k) puzzle, whose state space is a Cayley graph of Sn generated by a cyclic shift and a prefix reversal. We describe the mathematical framework, the algorithm, and its implementation, which combines a C++ hash-indexed state store with optional Vulkan GPU acceleration. The software is publicly available on CRAN.
Chinese Translation
我们提出了 cayleyR，这是一个用于通过检测 Cayley 图中的循环交集来解决置换拼图的 R 包。核心算法执行迭代双向搜索：从初始和目标置换状态出发，随机操作序列在对称群 Sn 的 Cayley 图中生成循环；它们的交集产生一条连接路径。当未找到直接交集时，基于距离的桥接选择缩小了差距，过程重复进行。该软件包针对 TopSpin(n,k) 拼图，其状态空间是由循环移位和前缀反转生成的 Sn 的 Cayley 图。我们描述了数学框架、算法及其实现，该实现结合了 C++ 哈希索引状态存储和可选的 Vulkan GPU 加速。该软件在 CRAN 上公开提供。
cs.AI / 10 / 2607.13220
Networked Intelligence: Active Shared Context Graphs for Human-AI Team Science
网络智能：用于人机团队科学的主动共享上下文图
Abstract
Most AI-for-science systems focus on scaling a single reasoning process through better models, larger context windows, long-horizon agentic execution, or digital co-scientists working with one principal user. However, challenging scientific problems are rarely solved by one reasoner alone. They are solved by teams whose members bring different priors, experimental backgrounds, tacit knowledge, and domain-trained intuitions. The open problem is therefore not only how to scale models, but how to cultivate networked intelligence: scaling the connections between humans and AI systems so that a result or hypothesis produced in one context reaches another person, agent, instrument, or robot that can act on it. We introduce Mycelium, an active shared workspace that automatically connects researchers and AI agents as a multi-user co-scientist. As human users and agents work, the system captures important observations and hypotheses, tracks how they relate to the team's evolving model, and routes them to the person or agent whose next decision they can inform. We evaluate Mycelium in its first empirical test, a biological multi-omics campaign in which routed shared context turned a local analytical finding into a cross-expert mechanistic constraint and ultimately into an experimental design. We also give networked intelligence a computational account as sparse conditional computation over distributed scientific contexts. This account distinguishes when a scaled standalone agent can match the network from when independent expertise and non-mergeable contexts make the network irreducible.
Chinese Translation
大多数用于科学的人工智能系统专注于通过更好的模型、更大的上下文窗口、长期的自主执行或与主要用户合作的数字共同科学家来扩展单一推理过程。然而，具有挑战性的科学问题很少由单一推理者解决。它们是由团队解决的，团队成员带来了不同的先验知识、实验背景、隐性知识和领域训练的直觉。因此，开放的问题不仅在于如何扩展模型，还在于如何培养网络智能：扩展人类与人工智能系统之间的连接，使得在一个上下文中产生的结果或假设能够传递给另一个可以对此采取行动的人、代理、仪器或机器人。我们引入了Mycelium，一个主动共享工作空间，能够自动连接研究人员和人工智能代理，作为多用户共同科学家。当人类用户和代理工作时，该系统捕捉重要的观察和假设，跟踪它们与团队不断发展的模型之间的关系，并将其路由到能够为下一步决策提供信息的人或代理。我们在Mycelium的首次实证测试中进行了评估，该测试是一项生物多组学活动，其中路由的共享上下文将局部分析发现转化为跨专家的机制约束，并最终转化为实验设计。我们还为网络智能提供了一个计算模型，作为对分布式科学上下文的稀疏条件计算。该模型区分了何时扩展的独立代理可以与网络匹配，以及何时独立的专业知识和不可合并的上下文使网络不可简化。
cs.AI / 11 / 2607.13230
AI-Native Insurance for Agentic AI: Pricing, Underwriting, and End-to-End Automation
面向自主人工智能的AI原生保险：定价、承保与端到端自动化
Abstract
Agentic AI introduces new insurance challenges because autonomous AI systems can make decisions, invoke tools, modify external environments, and interact with third-party services. This paper develops an AI-native mathematical framework for underwriting, pricing, and contract design for agentic AI deployments. A deployment is represented by a risk state that captures autonomy level, operational authority, permission exposure, governance maturity, and dependency concentration. The framework maps the risk state to event probabilities, loss severities, governance costs, premiums, deductibles, coverage allocation, and policy covenants, and formulates an optimization problem for insurance contract design under participation, profitability, and incentive compatibility constraints. The paper establishes structural properties of insurability, including characterization of an insurability region, monotone deterioration of feasibility with increasing exposure, and governance certification thresholds. Insurance is further interpreted as both an operational cost and a regulatory mechanism for AI deployment. A healthcare case study illustrates contract optimization, sensitivity analysis, and automated claims processing for agentic AI systems.
Chinese Translation
自主人工智能引入了新的保险挑战，因为自主AI系统可以做出决策、调用工具、修改外部环境并与第三方服务进行交互。本文开发了一个面向AI的数学框架，用于自主人工智能部署的承保、定价和合同设计。部署通过风险状态来表示，该状态捕捉自主级别、操作权限、许可暴露、治理成熟度和依赖集中度。该框架将风险状态映射到事件概率、损失严重性、治理成本、保费、自负额、覆盖分配和政策契约，并在参与性、盈利性和激励兼容性约束下，制定保险合同设计的优化问题。本文建立了可保险性的结构特性，包括可保险区域的特征描述、随着暴露增加可行性的单调恶化，以及治理认证阈值。保险进一步被解释为AI部署的运营成本和监管机制。一个医疗案例研究展示了自主人工智能系统的合同优化、敏感性分析和自动索赔处理。
cs.AI / 12 / 2607.13239
Cost-Optimal Foundation Model Deployment Portfolio for Transportation Management
交通管理的成本最优基础模型部署组合
Abstract
Foundation models, including large language models (LLMs) and vision-language models (VLMs), are increasingly used for transportation management center (TMC) tasks such as anomaly detection, incident reporting, and traveler information. Deploying multiple such models across TMC functions raises a portfolio question: which model should serve each function, in which deployment mode, and under what shared hardware budget? We formulate this as the Foundation Model Deployment Portfolio (FMDP) problem, a mixed-integer program minimizing total cost of ownership (TCO) subject to per-function quality, latency, and safety constraints over shared GPU capacity. We prove the problem NP-hard by reduction from the 0-1 knapsack problem and propose a polynomial-time greedy heuristic. In an illustrative case study with five TMC functions and 19 candidate (model, mode) pairs, FMDP identifies a mixed portfolio costing $34/mo (97% below the cheapest feasible all-closed-API baseline) by routing four functions to open-source APIs and the one function whose quality floor no open-source model meets to a closed API. Break-even analysis shows that on-premise GPU investment becomes reasonable only above approximately 309 vision queries/hour or if API prices double.
Chinese Translation
基础模型，包括大型语言模型（LLMs）和视觉语言模型（VLMs），在交通管理中心（TMC）的任务中越来越多地被使用，例如异常检测、事件报告和旅客信息。将多个此类模型部署在TMC功能中引发了一个组合问题：每个功能应使用哪个模型，以何种部署模式，以及在什么共享硬件预算下？我们将其表述为基础模型部署组合（FMDP）问题，这是一个混合整数规划，旨在最小化总拥有成本（TCO），同时满足每个功能的质量、延迟和安全约束，且在共享GPU容量下。我们通过从0-1背包问题的归约证明了该问题是NP难的，并提出了一种多项式时间贪心启发式算法。在一个包含五个TMC功能和19个候选（模型，模式）对的示例案例研究中，FMDP识别出一个混合组合，成本为每月34美元（比最便宜的可行全闭API基线低97%），通过将四个功能路由到开源API，而将一个质量底线没有任何开源模型满足的功能路由到闭合API。盈亏平衡分析显示，只有在每小时约309个视觉查询以上，或API价格翻倍时，现场GPU投资才变得合理。
cs.AI / 13 / 2607.13285
Harness Handbook: Making Evolving Agent Harnesses Readable,Navigable, and Editable
工具手册：使演化代理工具可读、可导航和可编辑
Abstract
The capability of a modern AI agent depends not only on its foundation model but also on its harness, which constructs prompts, manages state, invokes tools, and coordinates execution. As models, APIs, environments, and requirements evolve, the harness must be continually modified. Before such a change can be made, a developer or coding agent must identify all code locations that implement the target behavior. This is difficult because production harnesses are large, tightly coupled, and behaviorally distributed, while modification requests describe what the system should do and repositories are organized by files and modules. Code search, repository indexing, and long-context processing ease inspection, but still leave this behavior-to-code mapping to be recovered by hand. Behavior localization is therefore a central bottleneck in harness evolution. We introduce the Harness Handbook, a behavior-centric representation synthesized automatically from a harness codebase via static analysis and LLM-assisted structuring, linking each behavior to its corresponding source. We also introduce Behavior-Guided Progressive Disclosure (BGPD), which guides agents from high-level behaviors to relevant implementation details and verifies candidate locations against the current source. On diverse modification requests from two open-source harnesses, Handbook-Assisted planning improves behavior localization and edit-plan quality while using fewer planner tokens, with the largest gains on scattered sites, rarely executed paths, and cross-module interactions. Evolving complex agentic systems thus depends not only on generating edits, but also on determining where those edits should be made.
Chinese Translation
现代人工智能代理的能力不仅依赖于其基础模型，还依赖于其工具，这些工具构建提示、管理状态、调用工具并协调执行。随着模型、API、环境和需求的演变，工具必须不断修改。在进行此类更改之前，开发者或编码代理必须识别所有实现目标行为的代码位置。这是困难的，因为生产工具通常庞大、紧密耦合且行为分散，而修改请求描述了系统应执行的操作，代码库则按文件和模块组织。代码搜索、代码库索引和长上下文处理可以简化检查，但仍需手动恢复这种行为与代码的映射。因此，行为定位是工具演化中的一个核心瓶颈。我们引入了工具手册（Harness Handbook），这是一种以行为为中心的表示，通过静态分析和大型语言模型（LLM）辅助结构化自动从工具代码库中合成，链接每种行为及其对应的源代码。我们还引入了行为引导渐进披露（Behavior-Guided Progressive Disclosure，BGPD），该方法引导代理从高层行为到相关实现细节，并根据当前源代码验证候选位置。在来自两个开源工具的多样化修改请求中，手册辅助规划提高了行为定位和编辑计划的质量，同时使用了更少的规划令牌，尤其在分散位置、少量执行路径和跨模块交互上获得了最大的提升。因此，演化复杂的代理系统不仅依赖于生成编辑，还依赖于确定这些编辑应在哪里进行。
cs.AI / 14 / 2607.13292
Theory-Level Autoformalization: From Isolated Statements to Unified Formal Knowledge Bases
理论层面的自动形式化：从孤立语句到统一的形式知识库
Abstract
Autoformalization translates informal natural language into formal, machine-verifiable languages. While most work focuses on individual statements, real formalization efforts are inherently theory-level: they require an entire web of axioms, definitions, and lemmas before target theorems can even be stated. In this position paper, we argue for theory-level autoformalization: formalizing complete theories, including all their inter-dependencies, as structured libraries. We examine the significance of this shift, address alternative views, identify open challenges, and propose three promising paths forward. Our survey of autoformalization is available at https://github.com/marcusm117/Awesome-Autoformalization.
Chinese Translation
自动形式化将非正式自然语言转换为形式的、机器可验证的语言。尽管大多数研究集中于单个语句，但真正的形式化工作本质上是理论层面的：在目标定理能够被表述之前，需要一个完整的公理、定义和引理的网络。在这篇立场论文中，我们主张进行理论层面的自动形式化：将完整的理论形式化，包括所有的相互依赖关系，作为结构化的库。我们考察了这一转变的重要性，讨论了替代观点，识别了开放挑战，并提出了三条有前景的前进路径。我们的自动形式化调查可在 https://github.com/marcusm117/Awesome-Autoformalization 获取。
cs.AI / 15 / 2607.13344
EZSMT Version 3, Matured
EZSMT版本3，成熟
Abstract
Constraint Answer Set Programming (CASP) is a hybrid reasoning paradigm that combines Answer Set Programming (ASP) with Constraint Processing and Satisfiability Modulo Theories (SMT), enabling powerful declarative encodings of complex combinatorial search problems. This paper presents the design and implementation of EZSMTV3, an extensible SMT-based CASP framework that advances the translational approach to CASP solving. Building upon the foundation of the EZSMT+ system, EZSMTV3 introduces a more expressive input language, supports optimization via weak constraints, and offers foundations for streamlined integration of new constraint types. Rather than implementing custom search procedures, EZSMTV3 leverages state-of-the-art SMT solvers, such as CVC5, YICES, and Z3 to perform reasoning. The paper provides benchmarking results comparing EZSMTV3 with its CASP peers such as CLINGCON, CLINGO[DL], and CLINGO[LP], while showcasing its ability to handle mixed-domain constraints involving both integers and reals. The system provides a robust platform for future extensions and theoretical exploration within the CASP domain.
Chinese Translation
约束答案集编程（CASP）是一种混合推理范式，将答案集编程（ASP）与约束处理和模理论可满足性（SMT）相结合，使得复杂组合搜索问题的声明性编码变得强大。本文介绍了EZSMTV3的设计与实现，这是一个可扩展的基于SMT的CASP框架，推动了CASP求解的翻译方法。EZSMTV3在EZSMT+系统的基础上，引入了更具表现力的输入语言，支持通过弱约束进行优化，并为新约束类型的简化集成提供了基础。EZSMTV3并未实现自定义搜索过程，而是利用了最先进的SMT求解器，如CVC5、YICES和Z3来进行推理。本文提供了EZSMTV3与其CASP同行（如CLINGCON、CLINGO[DL]和CLINGO[LP]）的基准测试结果，同时展示了其处理涉及整数和实数的混合域约束的能力。该系统为未来在CASP领域的扩展和理论探索提供了一个强大的平台。
cs.AI / 16 / 2607.13396
Set-shifting Behavioral Test for Harnessed Agents
用于受控智能体的集合转移行为测试
Abstract
What happens to an LLM agent's tool choice when the reliable tool silently changes within an ongoing session? We borrow set-shifting from cognitive psychology to study how well agents adapt to hidden reliability shifts. Our benchmark mounts tool-skill libraries with redundancies, where many tools solve the same task but differ in hidden reliability. In our evaluation framework, a branched schedule shifts the reliable tool group at hidden boundaries and pairs every shift with a no-shift control. We find that agents, by default, settle on a small recurring routine within a few turns of each boundary, with call shares concentrating on a few discrete values after each reliability shift. We score the set-shifting accuracy for each agent trajectory: the joint probability of routing to the target tool group in every post-shift window. We test open-weight LLMs in an open-source agentic harness and find qualitatively distinct failure modes across the same set of routines. We also find that set framing, how the toolset presents the alternatives as competing or complementary, shifts the routing dynamics.
Chinese Translation
当可靠工具在进行中的会话中悄然变化时，LLM（大型语言模型）智能体的工具选择会发生什么？我们借鉴认知心理学中的集合转移理论，研究智能体如何适应隐藏的可靠性变化。我们的基准测试构建了具有冗余的工具技能库，其中许多工具可以解决相同的任务，但在隐藏的可靠性上存在差异。在我们的评估框架中，分支调度在隐藏边界处转移可靠工具组，并将每次转移与无转移控制进行配对。我们发现，智能体默认在每个边界的几个回合内会形成一个小的重复例程，调用分布在每次可靠性转移后集中在少数离散值上。我们为每个智能体轨迹评分集合转移的准确性：在每个转移后窗口中路由到目标工具组的联合概率。我们在一个开源的智能体框架中测试开放权重的LLM，并发现相同例程中存在定性不同的失败模式。我们还发现，集合框架，即工具集如何将替代方案呈现为竞争或互补，改变了路由动态。
cs.AI / 17 / 2607.13501
LAPO: Leave-One-Turn Attribution for Self-Generated Process Rewards in Multi-Turn Search Reasoning
LAPO：基于逐回合归因的自生成过程奖励在多回合搜索推理中的应用
Abstract
Reinforcement learning for multi-turn search reasoning typically relies on terminal outcome rewards, which cannot distinguish useful, redundant, and harmful intermediate interactions. We propose LAPO, a self-generated process-supervision method based on backward leave-one-turn attribution. For each search turn, LAPO replaces the turn and its retrieval observation with a fixed [DELETE] placeholder and measures the resulting change in the current policy's mean log-likelihood of the gold answer. This Answer-Likelihood Gain estimates the turn's contribution while preserving all downstream interactions, allowing early evidence to be evaluated in the complete reasoning context. LAPO further applies sign-consistency gating, retaining only normalized process advantages whose directions agree with their raw attribution scores. The method requires no additional reward model, teacher, verifier, or LLM-as-a-Judge. Across seven knowledge-intensive question-answering datasets with local retrieval, LAPO achieves an average exact-match score of 0.326, outperforming the strongest step-reward baseline, IGPO, by 0.053. Ablations show complementary benefits from backward attribution and sign-consistency gating, demonstrating that policy-derived retrospective attribution can provide effective process supervision for multi-turn search agents.
Chinese Translation
多回合搜索推理中的强化学习通常依赖于终端结果奖励，这无法区分有用、冗余和有害的中间交互。我们提出了LAPO，一种基于向后逐回合归因的自生成过程监督方法。对于每个搜索回合，LAPO用固定的[DELETE]占位符替换该回合及其检索观察，并测量当前策略的黄金答案的平均对数似然的变化。这种答案似然增益估计了该回合的贡献，同时保留所有下游交互，允许在完整的推理上下文中评估早期证据。LAPO进一步应用符号一致性门控，仅保留方向与其原始归因分数一致的标准化过程优势。该方法不需要额外的奖励模型、教师、验证者或LLM作为裁判。在七个知识密集型问答数据集上，LAPO在局部检索中实现了平均精确匹配分数0.326，超越了最强的步骤奖励基线IGPO，提升了0.053。消融实验显示，向后归因和符号一致性门控具有互补效益，证明了策略导出的回顾性归因可以为多回合搜索代理提供有效的过程监督。
cs.AI / 18 / 2607.13548
How Far Can Root Cause Analysis Go on Real-World Telemetry Data?
根本原因分析在真实世界遥测数据中的应用潜力有多大？
Abstract
Identifying root causes in production microservice failures requires reasoning over large-scale, multimodal telemetry spanning metrics, logs, and traces, a problem that has proved resistant to both classical and LLM-based approaches. The OpenRCA dataset exemplifies these challenges: it is large-scale, multimodal, and lacks detailed domain knowledge, and yields consistently low accuracy across all existing methods. We show that classical causal discovery methods and existing LLM-based multi-agent systems fail to reliably identify root causes on this benchmark, and present a Structured Multi-Agent RCA pipeline that substantially outperforms existing LLM-based and classical baselines, supporting both domain-knowledge and knowledge-free operating modes. To diagnose where failures originate, we introduce a reverse reasoning agent that, given the correct answer, identifies which signals in the extracted anomalies support it and determines whether Stage~1 had access to those signals, classifying each failure as Reasoning Gap (evidence present but unused) or Data Ambiguity (evidence genuinely absent). This analysis reveals that the required evidence is present in the vast majority of failures: the bottleneck is not data access but the agent's ability to reason over it correctly. We further introduce an automated rule mining pipeline that systematically extracts discrimination rules from reverse reasoning reports, reducing reliance on manual knowledge curation. Across all configurations, model reasoning capability and domain knowledge are the primary constraints: stronger models embed more domain expertise, and explicit knowledge injection partially compensates for this gap. Reasoning performance remains practically bounded even when evidence extraction is perfect: scaffold engineering and better data pipelines alone cannot close this gap; progress requires improvements at the model level.
Chinese Translation
在生产微服务故障中识别根本原因需要对大规模、多模态的遥测数据进行推理，这些数据涵盖了指标、日志和追踪，这一问题对经典方法和基于大语言模型（LLM）的方法均表现出抵抗力。OpenRCA 数据集 exemplifies 了这些挑战：它是大规模的、多模态的，并且缺乏详细的领域知识，导致所有现有方法的准确性始终较低。我们展示了经典因果发现方法和现有的基于 LLM 的多智能体系统在这一基准上无法可靠地识别根本原因，并提出了一种结构化多智能体 RCA 流水线，该流水线在性能上显著优于现有的基于 LLM 和经典的基线，支持领域知识和无知识操作模式。为了诊断故障的来源，我们引入了一个反向推理智能体，该智能体在给定正确答案的情况下，识别出提取的异常中支持该答案的信号，并确定第一阶段是否访问了这些信号，将每个故障分类为推理差距（证据存在但未使用）或数据模糊性（证据确实缺失）。这一分析揭示了所需证据在绝大多数故障中是存在的：瓶颈不在于数据访问，而在于智能体正确推理的能力。我们进一步引入了一种自动化规则挖掘流水线，该流水线系统性地从反向推理报告中提取区分规则，减少对手动知识整理的依赖。在所有配置中，模型推理能力和领域知识是主要限制：更强的模型嵌入了更多的领域专业知识，而显式知识注入在一定程度上弥补了这一差距。即使在证据提取完美的情况下，推理性能仍然受到实际限制：仅靠支架工程和更好的数据流水线无法弥补这一差距；进展需要在模型层面上进行改进。
cs.AI / 19 / 2607.13558
Multi-Agent Collaborative Reasoning with Tool-Augmented Evidence for Urban Region Profiling
基于工具增强证据的多智能体协作推理用于城市区域分析
Abstract
Urban region profiling constitutes a core problem in urban computing, supporting applications such as population estimation, economic assessment, and environmental monitoring. Existing methods typically formulate this task as multimodal representation learning, fusing heterogeneous urban data, e.g., satellite imagery, points of interest, textual descriptions, and 3D building information, into latent embeddings for prediction. However, these approaches are largely correlation-driven, assume cross-modal consistency, and rely on static pipelines, which limit their robustness in heterogeneous or unseen urban regions. We propose UrbanAgent, an agentic framework that reframes urban region profiling as a reasoning-driven inference problem. UrbanAgent instantiates an independent agent for each data modality and performs structured multi-agent collaborative reasoning to explicitly address cross-modal inconsistencies rather than absorbing them into a single representation. In addition, UrbanAgent extends indicator prediction as a closed-loop process of active evidence acquisition and iterative reasoning, enabling agents to verify uncertain inferences through tool-augmented retrieval of external knowledge optimized via reinforcement learning. Extensive experiments on global urban datasets for Carbon emissions, GDP, and Population estimation show that UrbanAgent consistently outperforms existing baselines, achieving an average improvement of 8.1% in R2, and exhibiting strong generalization performance in unseen-city settings.
Chinese Translation
城市区域分析是城市计算中的核心问题，支持人口估计、经济评估和环境监测等应用。现有方法通常将此任务表述为多模态表示学习，将异构城市数据（如卫星图像、兴趣点、文本描述和三维建筑信息）融合为潜在嵌入以进行预测。然而，这些方法主要是基于相关性驱动，假设跨模态一致性，并依赖静态管道，这限制了它们在异构或未见城市区域的鲁棒性。我们提出了UrbanAgent，一个将城市区域分析重新框定为推理驱动推断问题的智能框架。UrbanAgent为每种数据模态实例化一个独立的智能体，并进行结构化的多智能体协作推理，以明确解决跨模态不一致性，而不是将其吸收到单一表示中。此外，UrbanAgent将指标预测扩展为主动证据获取和迭代推理的闭环过程，使智能体能够通过强化学习优化的外部知识的工具增强检索来验证不确定的推断。在全球城市数据集上进行的关于碳排放、GDP和人口估计的广泛实验表明，UrbanAgent始终优于现有基线，R²平均提升8.1%，并在未见城市环境中表现出强大的泛化能力。
cs.AI / 20 / 2607.13562
AI advice suppresses people's willingness to say "I don't know", even when the advice is wrong and accuracy is incentivized
人工智能建议抑制人们说“我不知道”的意愿，即使建议是错误的且准确性受到激励
Abstract
Knowing when to say "I don't know" is fundamental to human judgment, yet AI assistants offer a fluent answer to almost any question. In five experiments (N = 3,132; four preregistered, one direct replication), participants answered difficult questions and could always decline to respond. We engineered the questions so that AI advice was wrong, separating AI use from its accuracy. Merely having access to AI nearly eliminated participants' willingness to suspend judgment, and this held whether the advice was actively requested or simply displayed. Consequently, participants answered more questions but were correct about a third as often as when AI was unavailable-yet their confidence nearly doubled. Incentivizing accuracy and penalizing inaccuracy led participants to seek and follow AI advice less, answer more accurately, and suspend judgment more often, though still far less than when AI was unavailable. As AI suggestions grow ubiquitous and unsolicited, they may not simply affect answer accuracy; they may even alter the metacognitive threshold at which people decide whether they know enough to answer.
Chinese Translation
知道何时说“我不知道”是人类判断的基础，但人工智能助手几乎可以流利地回答任何问题。在五个实验中（N = 3,132；四个预注册，一个直接复制），参与者回答了困难的问题，并且始终可以选择不作答。我们设计了这些问题，使得人工智能建议是错误的，从而将人工智能的使用与其准确性分开。仅仅拥有人工智能的访问权几乎消除了参与者暂停判断的意愿，无论建议是主动请求的还是仅仅显示的。因此，参与者回答了更多问题，但正确率仅为没有人工智能时的三分之一，然而他们的信心几乎翻倍。激励准确性并惩罚不准确性使参与者减少寻求和遵循人工智能建议的频率，提高了回答的准确性，并更频繁地暂停判断，尽管仍然远低于没有人工智能时的水平。随着人工智能建议的普遍存在和非请求性，它们可能不仅影响回答的准确性；甚至可能改变人们决定是否知道足够以回答问题的元认知阈值。
cs.AI / 21 / 2607.13594
SAFETY SENTRY: Context-Aware Human Intervention via EXECUTE-ASK-REFUSE Routing
安全哨兵：通过EXECUTE-ASK-REFUSE路由实现上下文感知的人类干预
Abstract
LLM agents act on real-world environments through tool calls, and a single misjudged action can cause irreversible harm. The standard safeguard is a guard model that labels each proposed action as safe or unsafe, but this binary view conflates two distinct decisions: whether the action is harmful in itself, and whether it is appropriate given the user's context. It also operates at the granularity of action categories rather than individual instances, producing routine interruptions that erode autonomy and train users to wave through the most consequential alerts. We reframe the problem as a per-instance three-way routing decision over {EXECUTE, ASK, REFUSE} and instantiate it with Safety Sentry, a lightweight guard model whose inference reduces to a single decoding call. A single decoding-time threshold lets one fixed checkpoint be re-positioned across deployments of differing risk tolerance without retraining. Safety Sentry outperforms a broad set of open-weight and frontier closed-source baselines on overall accuracy and safety-related recall, while controlling both directional error rates simultaneously.
Chinese Translation
大型语言模型（LLM）代理通过工具调用在现实世界环境中进行操作，而一次错误的判断可能导致不可逆转的伤害。标准的安全保障是一个守护模型，该模型将每个提议的行动标记为安全或不安全，但这种二元视角混淆了两个不同的决策：该行动本身是否有害，以及在用户的上下文中是否合适。它还在行动类别的粒度上进行操作，而不是针对单个实例，导致常规的干扰，侵蚀了自主性，并训练用户忽视最重要的警报。我们将问题重新构建为每个实例的三向路由决策，选项为{EXECUTE, ASK, REFUSE}，并通过安全哨兵（Safety Sentry）实现，该模型是一种轻量级的守护模型，其推理过程简化为单次解码调用。一个单一的解码时间阈值使得一个固定的检查点可以在不同风险容忍度的部署中重新定位，而无需重新训练。安全哨兵在整体准确性和与安全相关的召回率方面优于一系列开放权重和前沿闭源基准，同时同时控制方向性错误率。
cs.AI / 22 / 2607.13608
Automatic Ordinary Differential Equations Discovery For Biological Systems Using Large Language Model Powered Agentic System
基于大型语言模型驱动的自动常微分方程发现生物系统
Abstract
Automatic scientific discovery has long been a goal of computational scholars - a machine that can discover nature's secrets on its own, moving computational systems beyond data-fitting tools toward the generation and refinement of mechanistic models of the universe. Recent advances in symbolic regression (SR) and large-language-model (LLM)-based agents suggest that such systems can recover equations from data, incorporate domain priors, and automate parts of the research workflow. However, most existing approaches either focus on narrow equation-discovery benchmarks or broad end-to-end automation pipelines, while biological systems remain comparatively underexplored. Here, we introduce the MEDA system, an LLM- and SR-powered agentic framework for discovering ordinary-differential-equation (ODE) models of biological and biologically inspired dynamical systems. MEDA retrieves background knowledge, defines admissible variables, generates mechanistic constraints, proposes candidate ODEs, and fits and evaluates them. We evaluate it across canonical model retrieval, reasoning-based extrapolation to unseen variants, and open-ended discovery, with and without experimental data. Across these settings, MEDA recovered the correct state variables, achieved strong structural recovery in retrieval and extrapolation tasks, and produced biologically plausible discovery-oriented models. Ablation and robustness analyses show that knowledge-guided formalization and mechanistic constraints are load-bearing components, whereas numerical fitting alone can preserve trajectory-compatible but biologically incorrect equations.
Chinese Translation
自动科学发现一直是计算学者的目标——一种能够自主发现自然秘密的机器，将计算系统从数据拟合工具推进到宇宙机制模型的生成和完善。最近在符号回归（Symbolic Regression, SR）和基于大型语言模型（Large Language Model, LLM）的代理系统方面的进展表明，这类系统能够从数据中恢复方程，结合领域先验，并自动化研究工作流程的部分环节。然而，目前大多数现有方法要么专注于狭窄的方程发现基准，要么关注广泛的端到端自动化流程，而生物系统的研究相对较少。在此，我们介绍了MEDA系统，这是一个基于LLM和SR的代理框架，用于发现生物及生物启发的动态系统的常微分方程（Ordinary Differential Equation, ODE）模型。MEDA检索背景知识，定义可接受变量，生成机制约束，提出候选ODE，并进行拟合和评估。我们在经典模型检索、基于推理的对未见变体的外推以及开放式发现等方面对其进行了评估，既包括实验数据，也不包括实验数据。在这些设置中，MEDA恢复了正确的状态变量，在检索和外推任务中实现了强结构恢复，并生成了生物学上合理的发现导向模型。消融和鲁棒性分析表明，知识引导的形式化和机制约束是关键组成部分，而仅依靠数值拟合则可能保留与轨迹兼容但生物学上不正确的方程。
cs.AI / 23 / 2607.13618
STOCKTAKE: Measuring the Gap Between Perception and Action in LLM Agents with a Fair Oracle
STOCKTAKE：通过公平的Oracle测量LLM代理在感知与行动之间的差距
Abstract
LLM agents are increasingly evaluated on multi-week decision tasks in which the state that drives cost is never directly observed. On such tasks the final cost cannot say why an agent failed: it may have misread the world, or read it correctly and still failed to act (the knowing-doing gap). Existing evaluations cannot separate these two failures; their reference policies either read privileged information the agent never sees, or are missing altogether. We introduce STOCKTAKE, a 26-week supply-chain replenishment benchmark built as a factored partially observable Markov decision process with six hidden factor processes, designed so that a fair reference policy is computable: an exact Bayes filter per factor drives a rollout policy on the identical observation stream the agent receives. Scoring each run between a symptom-blind base-stock floor (0) and this oracle (1) yields a skill score, and grading each week's written rationale yields a stated-belief detection lag and a knowing-doing rate, so state estimation and control are measured separately. On fifty seeds with curated stress profiles, Claude Sonnet 5, GPT-5.4, DeepSeek-V4-Pro, and Grok 4.5 detect 84-88% of hidden failures, typically within a week of onset, yet span skill scores from 0.62 to -0.23: two of the four end below the symptom-blind floor while naming factors slightly faster than the two that beat it. The failure has two faces. Where stress persists, 34-43% of correctly diagnosed stress weeks still end in stockout for every model, a rate that partly reflects the severity of the weeks models notice. That rate also runs opposite to skill: the two models under the floor stock out least on diagnosed weeks, so under-response is only one face of the gap, and their traces point to the other, responses whose cost exceeds what they protect. STOCKTAKE measures both directions of that failure.
Chinese Translation
LLM代理在多周决策任务中的评估越来越多，这些任务的驱动成本状态从未被直接观察。在此类任务中，最终成本无法说明代理失败的原因：它可能误解了世界，或者正确理解却仍未采取行动（即知行差距）。现有的评估无法区分这两种失败；它们的参考策略要么读取了代理从未看到的特权信息，要么根本缺失。我们提出了STOCKTAKE，这是一个为期26周的供应链补货基准，构建为一个带有六个隐藏因子过程的分解部分可观察马尔可夫决策过程，旨在使公平的参考策略可计算：每个因子的精确贝叶斯滤波器驱动在代理接收的相同观察流上的滚动策略。在一个无症状的基础库存水平（0）和这个oracle（1）之间对每次运行进行评分，得出技能分数，并对每周的书面理由进行评分，得出声明信念的检测滞后和知行比率，从而分别测量状态估计和控制。在具有策划压力特征的五十个种子上，Claude Sonnet 5、GPT-5.4、DeepSeek-V4-Pro和Grok 4.5检测到84-88%的隐藏失败，通常在发作一周内，但技能分数跨度从0.62到-0.23：四个模型中有两个低于无症状基础水平，同时在命名因子方面略快于两个超越它的模型。失败有两个面向。在压力持续的情况下，34-43%的正确诊断压力周仍然以缺货告终，这一比例部分反映了模型注意到的周的严重性。该比例与技能呈相反关系：在诊断周中，低于基础水平的两个模型缺货最少，因此反应不足只是差距的一个面向，而它们的痕迹指向另一个，即其成本超过了它们所保护的成本。STOCKTAKE同时测量这种失败的两个方向。
cs.AI / 24 / 2607.13621
UESF-Bench: Benchmarking and Probing for Unified Embodied Seeking and Following
UESF-Bench：统一的具身寻求与跟随的基准测试与探测
Abstract
Language-guided human following is an important capability for embodied agents, but existing benchmarks typically assume that the target person is visible at the start of an episode. This setting simplifies the problem and overlooks a more realistic requirement: an agent often needs to first find a language-described target and then persistently follow that target in a dynamic environment. While recent work has started to study human search, existing settings are typically evaluated in task-specific scenarios and often rely on stronger prior knowledge of the environment. Moreover, they usually treat searching and following as separate tasks and still lack a unified benchmark for systematic evaluation. To address these limitations, we introduce the Unified Embodied Seeking and Following Benchmark (UESF-Bench), a large-scale and diverse benchmark for embodied human seeking and following. The benchmark requires agents to handle semantic-guided exploration, reliable behavior switching and recovery, and delayed identity grounding. To this end, we propose SeekFollow-VLA, a vision-language-action framework with a task-driven routing mechanism for latent phase inference and transition modeling between seeking and following. Experimental results show that SeekFollow-VLA achieves clear improvements over both single-head and dual-head baselines across single-person and multi-person environments, establishing a baseline for unified embodied seek-and-follow.
Chinese Translation
语言引导的人类跟随是具身智能体的重要能力，但现有基准通常假设目标人物在情节开始时是可见的。这种设置简化了问题，忽视了更现实的需求：智能体通常需要首先找到一个用语言描述的目标，然后在动态环境中持续跟随该目标。尽管最近的研究开始关注人类搜索，但现有设置通常在特定任务场景中进行评估，并且往往依赖于对环境的更强先验知识。此外，它们通常将搜索和跟随视为独立任务，仍然缺乏统一的基准进行系统评估。为了解决这些局限性，我们引入了统一的具身寻求与跟随基准（UESF-Bench），这是一个大规模且多样化的具身人类寻求与跟随基准。该基准要求智能体处理语义引导的探索、可靠的行为切换与恢复，以及延迟的身份定位。为此，我们提出了SeekFollow-VLA，这是一种具有任务驱动路由机制的视觉-语言-动作框架，用于潜在阶段推理和寻求与跟随之间的转移建模。实验结果表明，SeekFollow-VLA在单人和多人环境中相较于单头和双头基线均取得了显著提升，为统一的具身寻求与跟随建立了基线。
cs.AI / 25 / 2607.13655
Explaining Reinforcement Learning Agents via Inductive Logic Programming
通过归纳逻辑编程解释强化学习代理
Abstract
Explainable Reinforcement Learning (XRL) seeks to make Reinforcement Learning (RL) policies more transparent and interpretable, a key requirement in safety-critical and human-centric scenarios. However, it is mostly based on user studies, thus targeting the needs of a specific audience and lacking shared evaluation metrics. On the other hand, logic-based approaches within eXplainable Artificial Intelligence (XAI) provide compact, human-readable abstractions of decision-making. However, the systematic quantification of the explainability degree of logical representations remains an open problem. This work aims to advance the state of the art in XRL by introducing objective and planning-oriented metrics for policy explainability in RL settings. At the same time, it contributes to the field of logic for XAI by providing a principled way to quantify the explainability of logical rules, moving beyond common-sense assessments and simple propositional fragments. We employ Inductive Logic Programming (ILP) to extract symbolic representations of RL policies and define a novel set of explainability metrics, including activation rate, feature coverage, syntactic distance and semantic distance. These metrics quantify alignment between symbolic rules and agent behavior, the role of features in decision-making, and the evolution of policies during training and across agents in single and multi-agent RL. Experiments across different RL domains show that the proposed metrics highlight action-specific learning dynamics beyond global return, provide fine-grained insights into domain features beyond classical approaches for global feature importance estimation, and uncover coordination, specialization, and adaptation patterns in MARL. Moreover, they provide crucial insights for the transfer and generalization of action-specific policies.
Chinese Translation
可解释的强化学习（XRL）旨在提高强化学习（RL）策略的透明度和可解释性，这是在安全关键和以人为本的场景中一个关键要求。然而，目前的研究主要基于用户研究，因而针对特定受众的需求，缺乏共享的评估指标。另一方面，基于逻辑的方法在可解释人工智能（XAI）领域提供了紧凑且人类可读的决策抽象。然而，逻辑表示的可解释性程度的系统量化仍然是一个未解决的问题。本研究旨在通过引入针对RL环境中策略可解释性的客观和规划导向的指标，推动XRL的最新进展。同时，它为XAI领域的逻辑提供了一种原则性的方法，以量化逻辑规则的可解释性，超越常识评估和简单的命题片段。我们采用归纳逻辑编程（ILP）提取RL策略的符号表示，并定义了一组新的可解释性指标，包括激活率、特征覆盖率、句法距离和语义距离。这些指标量化了符号规则与代理行为之间的对齐程度、特征在决策中的作用，以及在单代理和多代理RL中策略在训练过程中的演变。不同RL领域的实验表明，所提出的指标突出了超越全局回报的特定动作学习动态，提供了对领域特征的细致洞察，超越了传统的全局特征重要性估计方法，并揭示了多智能体强化学习中的协调、专业化和适应模式。此外，它们为特定动作策略的迁移和泛化提供了重要的见解。
cs.AI / 26 / 2607.13679
When Bots Join the Team: Bot Adoption and the Institutional Fabric of Open-Source Software Projects
当机器人加入团队：机器人采纳与开源软件项目的制度结构
Abstract
AI agents are joining human teams, raising a basic question: when an automated agent becomes a regular participant, does group organization strengthen or weaken? We study this question in open-source software, where bots open pull requests, review code, and merge changes alongside people, leaving a public record of every interaction. Treating bots as participants rather than tools, we examine 2,991 GitHub projects for two years before and after each adopted its first bot. We measure three capabilities that institutional theory links to durable coordination - repeated engagement, social memory, and role differentiation - and two outcomes: conflict cascades and output distinctiveness. Bot adoption is followed by more repeated collaboration, greater recognition of specific bots in discussion, fewer conflict cascades, and more distinctive outputs. These changes cluster around adoption rather than accumulating gradually. Because we lack an untreated comparison group, we interpret the results as precisely timed associations, not causal effects. Two patterns are difficult for alternative explanations to account for: capabilities predict outcomes according to their function - coordination versus differentiation - rather than whether humans or bots provide them, and human-side capabilities account for the bot-conflict association but not the bot-distinctiveness association. The findings are consistent with a specific interpretation: predictable, rule-based agents can become part of a community's social infrastructure. The bot is the occasion; social organization is the mechanism.
Chinese Translation
人工智能代理正在加入人类团队，这引发了一个基本问题：当一个自动化代理成为常规参与者时，团队组织是增强还是削弱？我们在开源软件中研究这个问题，在这里，机器人打开拉取请求、审查代码并与人类一起合并更改，留下每次互动的公开记录。我们将机器人视为参与者而非工具，考察了2991个GitHub项目在采纳首个机器人之前和之后的两年。我们测量了与持久协调相关的三种能力——重复参与、社会记忆和角色区分——以及两个结果：冲突级联和输出独特性。机器人采纳后，重复合作增多，对特定机器人的讨论认可度提高，冲突级联减少，输出更具独特性。这些变化集中在采纳时刻，而非逐渐累积。由于缺乏未处理的对照组，我们将结果解读为精确时机的关联，而非因果效应。有两种模式难以用替代解释来说明：能力根据其功能（协调与区分）预测结果，而不是由人类或机器人提供，以及人类方面的能力解释了机器人与冲突之间的关联，但并未解释机器人与独特性之间的关联。这些发现与一种特定解释一致：可预测的、基于规则的代理可以成为社区社会基础设施的一部分。机器人是契机；社会组织是机制。
cs.AI / 27 / 2607.13705
AgentCompass: A Unified Evaluation Infrastructure for Agent Capabilities
AgentCompass：一个统一的智能体能力评估基础设施
Ding, Zichen, Ge, Jiaye, Jiang, Shufan, Chen, Kai, Li, Mo, Li, Qingqiu, Li, Zehao, Li, Zonglin, Liang, Tiaohao, Liu, Shudong, Ma, Zerun, Shang, Zixing, Tian, Wenhui, Wang, Zun, Wu, Liwei, Wu, Zhenyu, Xu, Jun, Yang, Bowen, Yuan, Dingbo, Zhang, Qi, Zhang, Songyang, Zhou, Peiheng, Zhu, Dongsheng
Abstract
As Large Language Models (LLMs) evolve into autonomous agents, the need for unified evaluation infrastructure becomes critical. However, current evaluation pipelines remain highly fragmented and tightly coupled, hindering reproducibility and causing redundant engineering. To address this, we introduce AgentCompass, an open-source, lightweight, and extensible infrastructure for evaluating LLM-based agents. AgentCompass organizes the evaluation process around three independent components, namely Benchmark, Harness, and Environment, thereby enabling flexible configurations without requiring the reimplementation of complex execution logic. Furthermore, it features a fault-tolerant asynchronous runtime and comprehensive trajectory analysis tools to transparently diagnose nuanced failure modes like reward-hacking. Natively supporting over 20 benchmarks across five capability dimensions, AgentCompass provides the community with a scalable and reproducible infrastructure for advancing agent research.
Chinese Translation
随着大型语言模型（LLMs）演变为自主智能体，统一评估基础设施的需求变得至关重要。然而，当前的评估流程仍然高度分散且紧密耦合，阻碍了可重复性并导致冗余的工程工作。为了解决这一问题，我们推出了AgentCompass，一个开源、轻量且可扩展的基础设施，用于评估基于LLM的智能体。AgentCompass围绕三个独立组件组织评估过程，即基准（Benchmark）、工具（Harness）和环境（Environment），从而实现灵活配置，而无需重新实现复杂的执行逻辑。此外，它具有容错的异步运行时和全面的轨迹分析工具，以透明地诊断诸如奖励操控（reward-hacking）等细微的失败模式。AgentCompass原生支持超过20个基准，涵盖五个能力维度，为社区提供了一个可扩展和可重复的基础设施，以推动智能体研究。
cs.AI / 28 / 2607.13716
CAVA: Canonical Action Verification and Attestation for Runtime Governance of Agentic AI Systems
CAVA：用于代理人工智能系统运行时治理的规范行动验证与证明
Abstract
Agentic AI systems increasingly act through heterogeneous runtimes: local coding hooks, SDK tools, browser automation, managed-agent traces, API gateways, and workflow engines. A single operational act such as publishing code, changing identity state, moving money, or exporting data may therefore be represented by many incompatible runtime records. This makes a basic governance question difficult to answer: what action was actually approved, what evidence binds the approval to execution, and can an independent verifier reproduce the same action identity later? This paper presents Canonical Action Verification and Attestation (CAVA), a runtime-semantics layer for converting heterogeneous agent activity into canonical runtime action objects. CAVA is positioned below Proof-Carrying Agent Actions (PCAA): PCAA defines the deployer-owned route-review-prove governance process, while CAVA defines the stable action object that process governs. The paper formalizes canonical action identity, semantic pattern detection, approval binding, receipt integrity, runtime-portable projection, and optional attestation substrates. We study a reference implementation through a 96-seed, 384-variant benchmark covering semantic equivalence, semantic separation, wrapper bypass, false-positive control, approval binding, receipt reproducibility, attestation tamper detection, runtime portability, semantic pattern detection, policy degradation, and Azure deployment drills. The contribution is a systems formulation of action-level canonicalization and policy-addressable semantic patterns as a necessary substrate for deployer-side AI governance.
Chinese Translation
代理人工智能系统越来越多地通过异构运行时进行操作：本地编码钩子、SDK工具、浏览器自动化、受管理代理的痕迹、API网关和工作流引擎。因此，单一的操作行为，如发布代码、改变身份状态、转移资金或导出数据，可能会由许多不兼容的运行时记录表示。这使得一个基本的治理问题难以回答：究竟批准了什么行动，什么证据将批准与执行绑定在一起，以及独立验证者能否在后续重现相同的行动身份？本文提出了规范行动验证与证明（CAVA），这是一个运行时语义层，用于将异构代理活动转换为规范运行时行动对象。CAVA 位于证明携带代理行动（PCAA）之下：PCAA 定义了部署者拥有的路线审查-证明治理过程，而 CAVA 定义了该过程所治理的稳定行动对象。本文对规范行动身份、语义模式检测、批准绑定、收据完整性、运行时可移植投影和可选证明基础进行了形式化研究。我们通过一个包含96个种子和384个变体的基准测试研究了一个参考实现，涵盖了语义等价、语义分离、包装器绕过、假阳性控制、批准绑定、收据可重现性、证明篡改检测、运行时可移植性、语义模式检测、政策降级和Azure部署演练。我们的贡献是将行动级规范化和政策可寻址语义模式的系统化表述作为部署者侧人工智能治理的必要基础。
cs.AI / 29 / 2607.13884
Experience Memory Graph: One-Shot Error Correction for Agents
经验记忆图：针对智能体的一次性错误修正
Abstract
Large Language Model (LLM) agents have shown remarkable capabilities in autonomous decision-making by generating sequential trajectories of states, actions, and observations. However, in complex, long-horizon tasks, these agents frequently suffer from compounding errors and struggle to recover from failures. Existing self-correction mechanisms rely on prompt-based reflection, which is inherently brittle, incurs heavy time and API costs due to iterative trial-and-error loops, and produces task-specific memory that may be hard to generalize to new scenarios. To address this, we propose Experience Memory Graph (EMG), a framework that reformulates agent failure recovery as a graph matching problem. At training time, we convert both failed exploration trajectories and successful expert trajectories into directed action decision graphs. By matching these graphs, we extract common subgraphs (successful workflows) and graph edit paths that explicitly indicate how to correct failures (e.g., which actions to add, delete, or relabel under a given observation), and store them in a memory graph with intra-task nodes and cross-task edges. At test time, EMG retrieves relevant insights and guides the agent in a single, loop-free execution. Experiments on ALFWorld and ScienceWorld show that EMG consistently outperforms state-of-the-art reflection baselines in success rate and average reward, while requiring no test-time trial-and-error.
Chinese Translation
大型语言模型（LLM）智能体在生成状态、动作和观察的序列轨迹方面展现了卓越的自主决策能力。然而，在复杂的长时间任务中，这些智能体常常遭遇累积错误，并难以从失败中恢复。现有的自我修正机制依赖于基于提示的反思，这种方法本质上脆弱，因迭代的试错循环而产生高昂的时间和API成本，并且生成的任务特定记忆可能难以推广到新场景。为了解决这个问题，我们提出了经验记忆图（Experience Memory Graph, EMG），一个将智能体失败恢复重新表述为图匹配问题的框架。在训练阶段，我们将失败的探索轨迹和成功的专家轨迹转换为有向动作决策图。通过匹配这些图，我们提取出共同的子图（成功工作流）和图编辑路径，明确指示如何修正失败（例如，在给定观察下需要添加、删除或重新标记哪些动作），并将其存储在一个具有任务内节点和任务间边的记忆图中。在测试阶段，EMG检索相关见解并指导智能体进行一次性、无循环的执行。在ALFWorld和ScienceWorld上的实验表明，EMG在成功率和平均奖励方面始终优于最先进的反思基准，同时在测试阶段无需试错。
cs.AI / 30 / 2607.13899
AIMO Interpretability Challenge
AIMO 可解释性挑战
Abstract
We propose the AIMO Interpretability Challenge, a competition on distinguishing robust from spurious reasoning in frontier mathematical language models based on the models' internal mechanisms. The challenge is motivated by a central limitation of standard reasoning benchmarks: strong final-answer accuracy does not reveal whether a model relies on stable reasoning mechanisms or exploits brittle reasoning shortcuts. Building on AI Mathematical Olympiad (AIMO) problems and submissions, together with resources from the Fields Model Initiative, the competition will provide (1) newly-published olympiad-level math reasoning problems and their symbolic representations, allowing generation of novel functional variants, (2) access to frontier reasoning models, and (3) assessments of models' adversarial robustness on these problems. Participants will use these resources, along with our computing infrastructure support, to develop methods for identifying which models solve problems robustly. Our competition will also create a new, open robustness benchmark and baseline systems, aiming to provide a lasting foundation for standard benchmarking in mathematical reasoning and interpretability. Scientifically, the competition connects interpretability and generalization research around a central question in AI research: can we determine if, and to what extent, the decision-making of frontier AI models is generalizable and thus, reliable?
Chinese Translation
我们提出了 AIMO 可解释性挑战，这是一个关于区分前沿数学语言模型中稳健推理与虚假推理的竞赛，基于模型的内部机制。该挑战的动机源于标准推理基准的一个核心局限性：强大的最终答案准确性并不能揭示模型是依赖于稳定的推理机制，还是利用脆弱的推理捷径。基于 AI 数学奥林匹克（AIMO）问题和提交的资源，以及来自菲尔兹模型倡议（Fields Model Initiative）的资源，本次竞赛将提供（1）新发布的奥林匹克级数学推理问题及其符号表示，允许生成新颖的功能变体，（2）对前沿推理模型的访问，以及（3）对这些问题上模型对抗鲁棒性的评估。参与者将利用这些资源，以及我们的计算基础设施支持，开发识别哪些模型能够稳健地解决问题的方法。我们的竞赛还将创建一个新的开放鲁棒性基准和基线系统，旨在为数学推理和可解释性领域的标准基准测试提供持久的基础。从科学角度来看，该竞赛将可解释性与泛化研究联系在一起，围绕人工智能研究中的一个核心问题：我们能否确定前沿人工智能模型的决策过程是否具有可泛化性，从而可靠？
cs.AI / 31 / 2607.13940
A Self-Evolving Agent for Longitudinal Personal Health Management
自我演化的长期个人健康管理代理
Abstract
Personal health management unfolds over repeated encounters, yet most health AI systems treat each request in isolation. We developed HealthClaw, an open-source agent architecture that updates support as a person's routines, preferences, measurements and risks change. It separates shared safety rules and medical knowledge from private longitudinal memory containing profile facts, reusable procedures and episodic traces. After each episode, induction determines what should update the profile, revise a procedure, remain episodic or be excluded. We evaluated HealthClaw with a synthetic year-long benchmark and nine 200-case biomedical tasks. Across 900 longitudinal support probes, answer accuracy increased from 0.2% with current-query prompting to 45.7% with HealthClaw, while prompt-side context exposure was 71.7% lower than with full-history prompting. In 100 privacy probes, HealthClaw produced higher privacy-aware answer quality and fewer unsafe disclosures than both baselines. Across the biomedical tasks, the mean absolute gain in the task-specific primary metric was 27.0 percentage points, and seven gains remained significant after false-discovery-rate correction. These offline benchmarks support governed, self-evolving memory for longitudinal personal health agents, although clinical effectiveness requires prospective evaluation. HealthClaw is publicly available at https://github.com/HC-Guo/HealthClaw.
Chinese Translation
个人健康管理是在重复的接触中展开的，但大多数健康人工智能系统将每个请求视为孤立的个体。我们开发了HealthClaw，一个开源代理架构，能够随着个人的日常活动、偏好、测量和风险的变化而更新支持。它将共享的安全规则和医学知识与包含个人档案事实、可重用程序和情节痕迹的私有长期记忆分开。在每个情节之后，归纳确定应更新哪些档案、修订哪些程序、保留哪些情节或排除哪些内容。我们通过一个合成的年度基准和九个200案例的生物医学任务评估了HealthClaw。在900个长期支持探测中，答案的准确性从当前查询提示的0.2%提高到使用HealthClaw的45.7%，而提示侧上下文的暴露比完全历史提示低71.7%。在100个隐私探测中，HealthClaw产生了更高的隐私意识答案质量和更少的不安全披露，相较于两个基线。对于生物医学任务，特定任务主要指标的平均绝对增益为27.0个百分点，七个增益在假发现率校正后仍然显著。这些离线基准支持受管控的自我演化记忆用于长期个人健康代理，尽管临床有效性需要前瞻性评估。HealthClaw可在https://github.com/HC-Guo/HealthClaw上公开获取。
cs.AI / 32 / 2607.14004
Do Agent Optimizers Compound? A Continual-Learning Evaluation on Terminal-Bench 2.0
代理优化器是否具有复合效应？在 Terminal-Bench 2.0 上的持续学习评估
Abstract
Most reported gains from agent-optimization methods are one-shot: an agent is optimized against a fixed benchmark and the resulting improvement is reported as if it were a stable property of the method. This does not test the setting that matters for deployed agents, where optimization is applied recursively as new failures and new tasks appear over time. The central question this raises is whether optimizer-driven gains compound: after an agent has been optimized once, can it be optimized again on newly arrived tasks without eroding the gains the first round produced? We study this question with a two-phase continual-learning evaluation built from hard tasks in Terminal-Bench 2.0, comparing three approaches to agent-harness optimization (GEPA, Meta Harness, and RELAI's Verifiable Continual Learning, RELAI-VCL) under identical optimization budgets. All three methods improve over the baseline agent in the conventional, static, single-phase setting. However, once new tasks are introduced, the methods diverge sharply: GEPA's optimized agent transfers below the unoptimized baseline, Meta Harness transfers well but fails to improve further once given a second optimization budget, and RELAI-VCL is the only method that both transfers positively to unseen tasks and continues improving after those tasks are folded into the optimization objective, reaching the highest pass rate at every evaluated stage and the highest lifelong average pass rate overall (76.4% vs. 66.0% for GEPA, 64.6% for Meta Harness, and 58.7% for the baseline). Our key observation was that optimization gains compounded only when regression control was built into the optimization loop, providing an inductive bias against shortcut solutions that fail to generalize.
Chinese Translation
大多数报告的代理优化方法的收益是一次性的：代理在固定基准上进行优化，所获得的改进被视为该方法的稳定特性。这并没有测试对于实际部署的代理而言重要的设置，即随着时间的推移，新的失败和新任务出现时，优化是递归应用的。由此产生的核心问题是，优化器驱动的收益是否会复合：在代理经过一次优化后，是否可以在新出现的任务上再次优化，而不损害第一次优化所产生的收益？我们通过在 Terminal-Bench 2.0 中构建的基于困难任务的两阶段持续学习评估来研究这个问题，比较三种代理优化方法（GEPA、Meta Harness 和 RELAI 的可验证持续学习 RELAI-VCL）在相同优化预算下的表现。所有三种方法在常规的静态单阶段设置中都优于基线代理。然而，一旦引入新任务，这些方法的表现差异显著：GEPA 优化后的代理在未优化的基线之下转移，Meta Harness 表现良好但在获得第二个优化预算后未能进一步改善，而 RELAI-VCL 是唯一一种在未见任务上正向转移并在这些任务被纳入优化目标后继续改善的方法，在每个评估阶段达到了最高的通过率，并且整体上达到了最高的终身平均通过率（76.4% 对比 GEPA 的 66.0%、Meta Harness 的 64.6% 和基线的 58.7%）。我们的关键观察是，只有当回归控制被纳入优化循环时，优化收益才会复合，从而对无法泛化的捷径解决方案提供了归纳偏见。
cs.AI / 33 / 2607.14044
AI-accelerated End-to-End Framework for Rapid Professional Upskilling
AI加速的端到端框架用于快速职业技能提升
Abstract
By 2030, 59 of every 100 workers will need reskilling or upskilling, yet the average time to close an enterprise skills gap grew from roughly 3 days in 2014 to 36 days in 2018. Most current frameworks accelerate single stages of upskilling programs and generally lack industry validation. We present an end-to-end framework that applies AI acceleration across five stages of knowledge acquisition, content development, content review and verification, teaching, and assessment development; with a strong focus on both production and learning efficiency. Three strong external signals validates the framework: the US National Association of State Boards of Accountancy reviewed and approved an upskilling program built on the framework for continuing-professional-education credits; 3 learners followed the program and passed the NVIDIA Certified Professional in Agentic AI exam in a significantly short amount of time, with 14 more in progress; the program's knowledge base supports complex downstream analysis such as the production of a robust 1,267 risk item dataset for managing multi-agent AI system risks.
Chinese Translation
到2030年，每100名工人中将有59人需要重新培训或提升技能，但企业技能差距的平均弥补时间从2014年的大约3天增长到2018年的36天。目前大多数框架仅加速技能提升程序的单个阶段，且普遍缺乏行业验证。我们提出了一个端到端框架，应用AI加速于知识获取、内容开发、内容审查与验证、教学和评估开发五个阶段，重点关注生产和学习效率。三个强有力的外部信号验证了该框架：美国国家会计委员会审查并批准了基于该框架的技能提升项目，以获得继续专业教育学分；3名学习者按照该项目学习，并在显著较短的时间内通过了NVIDIA认证的Agentic AI专业考试，另有14名学习者正在进行中；该项目的知识库支持复杂的下游分析，例如为管理多智能体AI系统风险生成一个强大的1,267个风险项数据集。
cs.AI / 34 / 2607.14046
Earthquaker-AI: A Retrieval-Augmented Generation Framework with Rubric-Based Assessment for Primary School Earthquake Education
Earthquaker-AI：一种基于评估标准的检索增强生成框架用于小学地震教育
Abstract
This paper presents Earthquaker-AI, a hybrid educational framework building upon a previously implemented educational robotics project by integrating a conversational AI assistant based on Retrieval-Augmented Generation. It aims to enhance earthquake preparedness and conscious action among primary-school students. The system extends the award-winning STEM project Earthquaker moving from mechanical simulation with Lego WeDo2 to cognitive and metacognitive processing. The robotics component uses Lego WeDo2 automation to simulate seismic response, letting students interact with sensors and actuators as tangible representations of protective actions. The assistant operates as a guided learning mechanism aligning student responses with safety guidelines, while providing rubric-based verbal feedback that supports self-regulated learning and calmness under emergency conditions. Earthquaker-AI follows a progressive learning trajectory aligned with cognitive development. In early grades, the focus is on basic recognition of safety actions through multiple-choice questions, assessed via a two-dimensional rubric. In middle grades, students identify correct action sequences through multiple-choice questions, evaluated via a three-axis rubric. In upper grades, the approach shifts to verbal production, requiring short written responses assessed via a four-dimensional rubric that includes clarity of expression. The dialogic module uses RAG to match student queries semantically with official guidelines, generating safe, accurate responses. Experimental evaluation shows high groundedness and accuracy, with a low hallucination rate. Overall, Earthquaker-AI combines hands-on engagement, information processing, and reflective practice. Combining robotics, rubrics, and AI promotes technological literacy, self-regulation, and responsible use of digital systems, contributing to early crisis-management skills.
Chinese Translation
本文介绍了Earthquaker-AI，这是一种混合教育框架，基于先前实施的教育机器人项目，通过整合基于检索增强生成的对话式人工智能助手来构建。其目的是增强小学学生的地震应急准备和意识行动。该系统扩展了获奖的STEM项目Earthquaker，从使用乐高WeDo2进行机械模拟转向认知和元认知处理。机器人组件利用乐高WeDo2自动化模拟地震反应，让学生与传感器和执行器互动，作为保护行动的具体表现。该助手作为一种引导学习机制，协调学生的反应与安全指南，同时提供基于评估标准的口头反馈，支持自我调节学习和在紧急情况下保持冷静。Earthquaker-AI遵循与认知发展相一致的渐进学习轨迹。在低年级阶段，重点是通过多项选择题基本识别安全行动，评估采用二维评估标准。在中年级阶段，学生通过多项选择题识别正确的行动顺序，评估采用三维评估标准。在高年级阶段，方法转向口头表达，要求学生提供简短的书面回答，评估采用四维评估标准，其中包括表达的清晰度。对话模块使用RAG将学生查询与官方指南在语义上匹配，生成安全、准确的响应。实验评估显示出高的基础性和准确性，且幻觉率低。总体而言，Earthquaker-AI结合了实践参与、信息处理和反思实践。结合机器人技术、评估标准和人工智能促进了技术素养、自我调节和数字系统的负责任使用，为早期危机管理技能的培养做出了贡献。
cs.AI / 35 / 2607.14049
Deep Interaction: An Efficient Human-AI Interaction Method for Large Reasoning Models
深度互动：一种高效的人机交互方法用于大型推理模型
Abstract
The emergence of Chain-of-Thought (CoT) reasoning has significantly enhanced the ability of large language models (LLMs) to tackle complex, multi-step tasks. However, when errors occur, current interaction approaches typically involve re-generating another response that may make mistakes again, or users laboriously flag the faulty step in follow-up turns that may get responses followed by similar errors recurring. To address this issue, we propose an efficient human intervention mechanism for precisely correcting reasoning errors in LLMs, termed Deep Interaction. Our approach enables direct editing of the original response, allowing erroneous parts to be corrected while preserving accurate reasoning steps. We refine the edited CoT into a distilled prompt, which then steers the LLM along the corrected reasoning path. Experimental results show that our method achieves over a 25% improvement in correction success rate and reduces token usage by approximately 40% on STEM tasks reasoning compared to baseline approaches.
Chinese Translation
链式思维（Chain-of-Thought, CoT）推理的出现显著增强了大型语言模型（Large Language Models, LLMs）处理复杂多步骤任务的能力。然而，当出现错误时，目前的交互方法通常涉及重新生成另一个可能再次出错的响应，或者用户在后续回合中费力地标记错误步骤，这可能导致响应 后面出现类似的错误。为了解决这个问题，我们提出了一种高效的人为干预机制，用于精确纠正LLMs中的推理错误，称为深度互动（Deep Interaction）。我们的方法允许直接编辑原始响应，纠正错误部分，同时保留准确的推理步骤。我们将编辑后的CoT精炼为一个浓缩提示（distilled prompt），然后引导LLM沿着纠正后的推理路径进行推理。实验结果表明，我们的方法在纠正成功率上提高了超过25%，并且在STEM任务推理中相比基线方法减少了约40%的令牌使用。
cs.CL / 1 / 2607.13035
FixItFlow: Automated Troubleshooting Guide Generation from Cloud Incidents
FixItFlow：基于云事件的自动故障排除指南生成
Abstract
Cloud services experience frequent incidents that require rapid diagnosis and resolution. Troubleshooting guides help engineers respond consistently, but creating them manually is labor-intensive, resulting in incomplete coverage and outdated documentation. We present FixItFlow, an automated system that generates troubleshooting guides from historical incident data using large language models. The system extracts diagnostic patterns from engineer actions, synthesizes structured guides with verified commands, and enforces strict validation to prevent fabricated content. In our evaluation with 26 engineers, generated guides achieved 61.5\% positive ratings for clarity and demonstrated a 2.3x reduction in mitigation time for incidents with associated guides. These results indicate that automated guide generation can improve incident response while reducing documentation burden on engineering teams.
Chinese Translation
云服务经常发生需要快速诊断和解决的事件。故障排除指南帮助工程师保持一致的响应，但手动创建这些指南劳动密集，导致覆盖不全和文档过时。我们提出了FixItFlow，一个自动化系统，利用大型语言模型从历史事件数据中生成故障排除指南。该系统从工程师的操作中提取诊断模式，合成包含经过验证命令的结构化指南，并严格执行验证以防止虚假内容的产生。在我们与26名工程师的评估中，生成的指南在清晰度上获得了61.5%的正面评价，并且在与指南相关的事件中，缓解时间减少了2.3倍。这些结果表明，自动化指南生成可以改善事件响应，同时减轻工程团队的文档负担。
cs.CL / 2 / 2607.13036
Ask Before You Diagnose: Safe-Psych, a Sequential Evaluation Benchmark for LLMs in Psychiatry
在诊断之前询问：Safe-Psych，一个用于评估精神病学中大型语言模型的顺序评估基准
Abstract
Large language models (LLMs) are increasingly used for decision support in healthcare, but clinical evidence is often incomplete or evolving. When the available information is insufficient to support a reliable answer, models should request clarification or abstain rather than provide unsupported responses. Existing medical benchmarks, however, typically assume that complete information is available upfront. We introduce Safe-Psych, a sequential benchmark for evaluating how LLMs handle evolving diagnostic uncertainty in clinical psychiatry. Safe-Psych contains over 1,000 real-world psychiatric clinical notes segmented to simulate incremental evidence disclosure, with psychiatrist-derived action labels at each stage: DIAGNOSE, CLARIFY, or ABSTAIN. We evaluate multiple state-of-the-art LLMs in full-information and sequential settings. Our findings show that capability does not ensure calibration: even strong models struggle under incomplete clinical information, with under-abstention exceeding 60% for most models and safety-aware prompting reducing premature commitment only by shifting errors toward excessive abstention. In sequential evaluation, models frequently diagnose before sufficient evidence is available and rarely seek clarification unless explicitly prompted; these premature diagnoses are less accurate than on-time diagnoses. Overall, Safe-Psych reveals a limitation across the evaluated models: recognizing when clinical evidence is incomplete and additional information is needed. We release Safe-Psych to support research on improving LLM safety in healthcare.
Chinese Translation
大型语言模型（LLMs）在医疗保健中的决策支持中越来越多地被使用，但临床证据往往是不完整或不断变化的。当可用信息不足以支持可靠答案时，模型应该请求澄清或选择不作答，而不是提供没有支持的回应。然而，现有的医学基准通常假设在开始时就有完整的信息。我们引入了Safe-Psych，这是一个用于评估LLMs如何处理临床精神病学中不断变化的诊断不确定性的顺序基准。Safe-Psych包含超过1000个真实世界的精神病临床记录，这些记录被分段以模拟逐步证据披露，并在每个阶段附有精神科医生制定的行动标签：诊断（DIAGNOSE）、澄清（CLARIFY）或放弃（ABSTAIN）。我们在完整信息和顺序设置中评估了多种最先进的LLMs。我们的研究结果表明，能力并不确保校准：即使是强大的模型在不完整的临床信息下也会遇到困难，大多数模型的放弃不足超过60%，而安全意识提示仅通过将错误转向过度放弃来减少过早承诺。在顺序评估中，模型在证据不足时经常进行诊断，且除非明确提示，否则很少寻求澄清；这些过早的诊断比及时的诊断准确性更低。总体而言，Safe-Psych揭示了被评估模型的一个局限性：识别临床证据何时不完整以及何时需要额外信息。我们发布Safe-Psych以支持改善医疗保健中LLM安全性的研究。
cs.CL / 3 / 2607.13044
The Perplexity Trap: When Patent Law Makes Human Writing Look Like AI
困惑陷阱：当专利法使人类写作看起来像人工智能
Abstract
The European Patent Office (EPO) reported record filings in 2025, and the 2026 EPO Guidelines hold applicants strictly responsible for LLM-assisted content under Article 83 and Rule 42, creating pressure to triage suspected AI-generated patent text. Two constraints make this hard. First, realistic prosecution settings often have only consumer GPUs with about 8 GB VRAM, not datacenter-class scoring stacks. Second, Article 84 of the European Patent Convention requires claims to be clear and concise, pushing human drafting onto the same low-perplexity, low-burstiness manifold that LLMs occupy. We benchmark three open-source zero-shot detectors on 500 granted EPO H04 telecom patents versus 500 LLM-generated counterparts using five prompting strategies, all under the consumer hardware envelope. At claim level, all detectors exceed 60 percent false-positive rate: Binoculars 78.3 percent, Fast-DetectGPT 61.3 percent, DetectGPT 80.5 percent. The failure persists under Qwen2.5-3B-Instruct regeneration, LoRA-adapted Pythia-2.8B scoring heads, cross-IPC replication on A61K, C07D, and F03D (mean FPR 84.6 percent), and H100 re-evaluation with published Falcon-7B and GPT-J-6B heads, arguing the issue is structural rather than substitute-model capacity. A seven-feature linguistic-complexity logistic regression reaches 74.0 percent accuracy at 28.1 percent FPR, a 13 percentage-point gain over a perplexity-only baseline at a comparable operating point, without using likelihood at inference and within the same hardware budget.
Chinese Translation
欧洲专利局（EPO）在2025年报告了创纪录的申请数量，而2026年EPO指南要求申请人对根据第83条和第42条规则的LLM辅助内容承担严格责任，这给怀疑是人工智能生成的专利文本带来了筛选压力。两个限制因素使这一过程变得困难。首先，现实的审查环境通常只有约8 GB显存的消费级GPU，而没有数据中心级别的评分堆栈。其次，欧洲专利公约第84条要求权利要求必须清晰简洁，这将人类撰写推向与LLM相同的低困惑度、低突发性流形。我们在500个已授予的EPO H04电信专利与500个LLM生成的对应专利上，使用五种提示策略对三种开源零-shot检测器进行了基准测试，所有测试均在消费级硬件环境下进行。在权利要求层面，所有检测器的假阳性率均超过60%：Binoculars 78.3%、Fast-DetectGPT 61.3%、DetectGPT 80.5%。在Qwen2.5-3B-Instruct再生、LoRA适配的Pythia-2.8B评分头、A61K、C07D和F03D的跨IPC复制（平均假阳性率84.6%）以及使用已发布的Falcon-7B和GPT-J-6B评分头的H100重新评估下，这一失败依然存在，表明问题是结构性的，而非替代模型能力的问题。一个包含七个特征的语言复杂性逻辑回归模型在28.1%的假阳性率下达到了74.0%的准确率，比在可比操作点上的仅基于困惑度的基线提高了13个百分点，且在推理时未使用似然，并且在相同的硬件预算内。
cs.CL / 4 / 2607.13158
Do LLMs Need Architectural Changes for Simultaneous Speech Translation? A Prefix-to-Prefix Data Driven Approach
大型语言模型是否需要架构变更以实现同步语音翻译？一种基于前缀到前缀的数据驱动方法
Abstract
Simultaneous speech translation (SimulST) requires incremental translation under strict latency constraints, yet remains challenging for decoder-only LLM systems due to limited context and cross-lingual reordering. Recent approaches often introduce architectural changes or explicit read/write policies to control output timing, which can be brittle in conversational speech where segmentation boundaries are ambiguous. We present a simple data-driven alternative: fixed-length chunks for cumulative streaming decoding with a rewind-based committed prefix, and teacher-labeled prefix-to-prefix (P2P) targets with bounded waiting for fine-tuning, yielding CSSEL-P2P, where CSSEL is our proposed chunked streaming speech encoder LLM. In our in-house conversational speech evaluation, CSSEL-P2P improves streaming quality by +1.54 COMETKiwi over the CSSEL streaming baseline at comparable latency (+0.15s Average Lagging), suggesting effective SimulST without architectural changes via P2P supervision.
Chinese Translation
同步语音翻译（SimulST）要求在严格的延迟限制下进行增量翻译，但由于上下文有限和跨语言重排序，解码器仅的LLM系统仍然面临挑战。近期的方法通常引入架构变更或显式的读/写策略来控制输出时机，但在对话语音中，分段边界模糊，这可能导致脆弱性。我们提出了一种简单的数据驱动替代方案：使用固定长度的块进行累积流解码，并结合基于回退的已承诺前缀，以及带有有限等待的教师标记前缀到前缀（P2P）目标进行微调，形成CSSEL-P2P，其中CSSEL是我们提出的分块流语音编码器LLM。在我们内部的对话语音评估中，CSSEL-P2P在可比延迟下（+0.15秒平均滞后）相较于CSSEL流基线提高了+1.54 COMETKiwi的流质量，表明通过P2P监督实现有效的SimulST而无需架构变更。
cs.CL / 5 / 2607.13162
What Models Express, Suppress, and Resist: Auditing Open-Weight LLMs with Persona Vectors
模型的表达、抑制与抵抗：使用个性向量审计开放权重的大型语言模型
Abstract
What a language model will and will not do is largely set during post-training, but which behaviors it expresses, hides, or resists is not revealed by prompting alone. Persona vectors, behavioral directions in activation space, can probe this organization, but prior work covers only a handful of traits. We present the first systematic application of persona vectors at this scale, compiling a 53-trait inventory across four behaviorally distinct domains and labeling every trait in two open-weight models as natural (expressed at baseline), steerable latent but amplifiable, or intractable (resistant to standard extraction). Both models default to helpful, task-oriented behavior: all nine agentic traits are natural, and their default clinician behavior matches a board-certified psychologist's independent desirability judgments on 16 of 17 traits. Steering produces its largest gains on traits these defaults exclude: hyperbole, hallucination, and sycophancy. The same asymmetry holds across all 171 generic-trait pairs: two steerable traits can collapse the composition, but pairs involving a default never do. Where standard extraction fails on a trait like "evil," a vector transferred from a fine-tuned variant still recovers it, with the residual refusals appearing inside the model's chain-of-thought. Persona vectors are most informative not as a set of controls but as a probe of behavioral organization.
Chinese Translation
语言模型的行为在后训练阶段基本确定，但其表达、隐藏或抵抗的行为仅通过提示并不能完全揭示。个性向量作为激活空间中的行为方向，可以探测这种组织结构，但之前的研究仅涵盖了少数特征。我们首次在这一规模上系统性地应用个性向量，编制了一个涵盖四个行为上明显不同领域的53个特征的清单，并将每个特征在两个开放权重模型中标记为自然（在基线下表达）、可引导的潜在但可放大的，或不可处理的（对标准提取具有抵抗性）。这两个模型默认表现出有帮助的、以任务为导向的行为：所有九个主动特征都是自然的，并且它们的默认临床行为与一位获得认证的心理学家对17个特征中16个的独立可取性判断相匹配。引导在这些默认特征所排除的特征上产生了最大的增益：夸张、幻觉和谄媚。在所有171对通用特征中，这种不对称性依然成立：两个可引导的特征可以压缩组合，但涉及默认特征的对从未如此。当标准提取在“邪恶”等特征上失败时，从微调变体转移的向量仍然能够恢复该特征，残余的拒绝出现在模型的思维链中。个性向量最有价值的不是作为一组控制变量，而是作为探测行为组织的工具。
cs.CL / 6 / 2607.13164
Text2Sign: A Single-GPU Diffusion Baseline for Text-to-Sign Language Video Generation
Text2Sign：一种基于单个GPU的文本到手语视频生成的扩散基线
Abstract
Sign language is a primary communication channel for millions of Deaf and hard-of-hearing people, yet text-to-signer video generation remains costly because video diffusion models are expensive to train and evaluate. This paper presents Text2Sign, a text-conditioned diffusion model for short sign-language clips that runs on a single NVIDIA L4 GPU. It combines a frozen vision-language text encoder with a 3D encoder-decoder and factorized spatiotemporal attention to reduce the cost of full-video attention while preserving motion coherence. We compare convolution-only and transformer-style backbones, frozen pretrained and task-specific text encoders, and factorized versus full attention. On a signer-disjoint How2Sign split, the best short-run ablation reaches a validation loss of 0.0648, while a longer-run checkpoint reaches 0.00999. On a compact evaluation slice, the latter achieves an SSIM of $0.2403 \pm 0.0238$, a PSNR of $15.11 \pm 0.42$ dB, and temporal consistency of $1.0000 \pm 0.0000$ using 8-step DDIM sampling with a guidance scale of 5.0. It generates a 32-frame, $64 \times 64$ clip in 12.60 seconds, or 2.54 frames per second, with peak inference memory of 3.12 GB. A held-out denoising audit shows only weak prompt sensitivity: removing text increases late-timestep loss from 0.9875 to 0.9891, while shuffled prompts perform similarly to correct prompts. Frozen text conditioning therefore improves short-budget validation loss, but prompt-specific separation remains limited. The system is restricted to low-resolution, short clips and lacks expert linguistic evaluation, so it should be viewed as a single-GPU research baseline rather than a complete sign-language production system. Code is available at https://github.com/xiaruize0911/text2sign.
Chinese Translation
手语是数百万聋人和重听者的主要沟通渠道，但文本到手语视频生成仍然成本高昂，因为视频扩散模型的训练和评估都非常耗费资源。本文提出了Text2Sign，一种基于文本条件的扩散模型，用于短手语片段，能够在单个NVIDIA L4 GPU上运行。该模型结合了一个冻结的视觉-语言文本编码器、一个3D编码器-解码器以及分解的时空注意力，以降低全视频注意力的成本，同时保持运动一致性。我们比较了仅卷积和变换器风格的骨干网络、冻结的预训练和任务特定的文本编码器，以及分解注意力与全注意力的效果。在一个与手语者不重叠的How2Sign数据集划分上，最佳的短期消融实验达到了0.0648的验证损失，而较长期的检查点达到了0.00999。在一个紧凑的评估片段上，后者的结构相似性指数（SSIM）为$0.2403 imes 0.0238$，峰值信噪比（PSNR）为$15.11 imes 0.42$ dB，时间一致性为$1.0000 imes 0.0000$，使用8步DDIM采样和5.0的引导比例。它在12.60秒内生成了一个32帧、$64 imes 64$的片段，帧率为2.54帧每秒，峰值推理内存为3.12 GB。一次保留的去噪审计显示出仅有微弱的提示敏感性：移除文本会使后期时间步的损失从0.9875增加到0.9891，而随机打乱的提示与正确提示的表现相似。因此，冻结的文本条件改善了短预算验证损失，但提示特定的分离仍然有限。该系统仅限于低分辨率、短片段，并缺乏专家语言评估，因此应被视为单个GPU的研究基线，而非完整的手语生成系统。代码可在https://github.com/xiaruize0911/text2sign获取。
cs.CL / 7 / 2607.13189
RAGthoven at SemEval-2026 Task 1: A Multi-Stage Pipeline Walks Into a Benchmark and Barely Clears the Bar
RAGthoven在SemEval-2026任务1中的表现：一个多阶段管道走进基准测试，勉强达标
Abstract
We present RAGthoven, our system for SemEval-2026 Task 1 (MWAHAHA), Subtask A (multilingual constrained humor generation in English, Spanish, and Chinese). RAGthoven decomposes creative text generation into a multi-stage large language model (LLM) pipeline (Planner, Best-of-N Writer, Reflector for self-critique, LLM-as-a-judge Judge) grounded in computational humor theory (Benign Violation Theory, Script-based Semantic Theory of Humor) and refined across ten experiments. In our final configuration, we augment the Planner with retrieval-augmented generation (RAG) from a curated joke corpus, seeding generation with diverse joke mechanisms. We also evaluate two agentic variants -- ReAct-style sequential tool-calling (Exp09) and autonomous multi-branch orchestration (Exp10) -- that expose the same four stages with a deterministic ConstraintAudit checker. Across four frontier models on a held-out 12-instance English sample, neither agentic variant produced outputs we judged superior to the non-agentic pipeline despite substantially higher tool-call budgets. RAGthoven shares Rank 1 with the Gemini 2.5 Flash baseline in all three languages, with overlapping organizer-reported confidence intervals. In Spanish, it leads the baseline by 42 raw Elo points (1182 vs. 1140), while in English (1045 vs. 1081) and Chinese (1045 vs. 1053) the baseline holds the higher raw rating within the same statistical tie. Together, these results suggest language-dependent diminishing returns from elaborate multi-stage prompt engineering and agentic scaffolding once a strong frontier model is in the loop.
Chinese Translation
我们介绍了RAGthoven，这是我们在SemEval-2026任务1（MWAHAHA）中的系统，子任务A（英语、西班牙语和中文的多语言受限幽默生成）。RAGthoven将创造性文本生成分解为一个多阶段的大型语言模型（LLM）管道（规划者、最佳写作者、反思者用于自我批评、LLM作为评审的评审者），该管道基于计算幽默理论（良性侵犯理论、基于脚本的幽默语义理论）并经过十次实验的精炼。在我们的最终配置中，我们通过从一个策划的笑话语料库中引入检索增强生成（RAG）来增强规划者，以多样的笑话机制作为生成的种子。我们还评估了两种代理变体——ReAct风格的顺序工具调用（Exp09）和自主多分支编排（Exp10），这两者都暴露出相同的四个阶段，并配备了确定性的约束审计检查器。在一个保留的12个实例的英语样本上，四个前沿模型中，没有任何代理变体产生我们认为优于非代理管道的输出，尽管其工具调用预算显著更高。RAGthoven在所有三种语言中与Gemini 2.5 Flash基线共享第一名，并且重叠的组织者报告的置信区间。在西班牙语中，它领先基线42个原始Elo点（1182对1140），而在英语（1045对1081）和中文（1045对1053）中，基线在同一统计平局中保持更高的原始评分。综合来看，这些结果表明，一旦强大的前沿模型参与其中，复杂的多阶段提示工程和代理支架的语言依赖性收益递减。
cs.CL / 8 / 2607.13205
Adaptive Filtering of the KV Cache: Diagnosing and Correcting Structural-Role Bias in LLM Inference
KV缓存的自适应过滤：诊断和纠正LLM推理中的结构角色偏差
Abstract
Attention-based KV cache eviction (H2O and its descendants) compresses the memory-constrained state of a long-context model by ranking tokens on accumulated attention mass, treated here as signal energy, and keeping the heaviest. On schema-dense input streams such as nested JSON, this score acts as a non-stationary filter that disproportionately retains noise: a non-content sink role (delimiters or whitespace) carries an order of magnitude more energy than any content role, and structural KEY tokens are over-retained at roughly 1.8x the rate of the answer-carrying VALUE tokens, collapsing exact-match accuracy from 88% to 0% at a 5% budget as the signal-to-noise ratio of the retained state degrades. A counterfactual experiment establishes that suppressing KEY tokens is the best deployable filter. Our retraining-free, role-conditional allocation over SnapKV's windowed score, governed by a single tuned hyperparameter, closes 63-98% of the H2O gap at sub-20% budgets and, at higher budgets, modestly matches or exceeds full-cache accuracy -- a small, seed-sensitive denoising effect (borderline significant at B=0.50; not distinguishable from zero at B=0.30 over four seeds). A 15 MB linear role probe supplies these labels at negligible inference cost, though matching parser-level downstream accuracy remains open.
Chinese Translation
基于注意力的KV缓存驱逐（H2O及其后代）通过对累积注意力质量进行排名来压缩长上下文模型的内存受限状态，这里将其视为信号能量，并保留能量最大的标记。在像嵌套JSON这样的模式密集输入流中，这一评分作为一种非平稳过滤器，过度保留了噪声：非内容的接收角色（分隔符或空白）携带的能量比任何内容角色高一个数量级，而结构性KEY标记的保留率约为承载答案的VALUE标记的1.8倍，导致在5%的预算下，保留状态的信噪比下降，使得精确匹配准确率从88%降至0%。一个反事实实验表明，抑制KEY标记是最佳的可部署过滤器。我们的无重训练、基于角色条件的分配方法在SnapKV的窗口评分上，通过一个调优的超参数，能够在低于20%的预算下关闭63-98%的H2O差距，并且在更高预算下，适度匹配或超过全缓存的准确率——这是一种小的、对种子敏感的去噪效果（在B=0.50时接近显著；在B=0.30时在四个种子上与零不可区分）。一个15 MB的线性角色探针以微不足道的推理成本提供这些标签，尽管匹配解析器级下游准确率仍然是一个开放问题。
cs.CL / 9 / 2607.13248
GSM-Plus-BN: A Perturbation-Based Benchmark for Bangla Mathematical Reasoning in Large Language Models
GSM-Plus-BN：一种基于扰动的孟加拉数学推理基准测试
Abstract
The evaluation of mathematical reasoning in large language models (LLMs) has predominantly focused on high-resource languages like English. This has created a significant barrier to the equitable development and deployment of AI in linguistically diverse regions such as Bangladesh, where over 230 million people speak Bengali. Despite this global significance, there has been minimal prior work on mathematical reasoning in Bengali and no existing research that systematically benchmarks a perturbated Bengali mathematical dataset, leaving a critical void in assessing model robustness and true comprehension beyond pattern recognition. This study addresses this gap by introducing GSM-Plus-BN, a novel perturbated Bengali mathematical dataset derived from the English GSM-Plus benchmark and verified by human translators. We evaluate six open-source LLMs Qwen3-32B, Llama-3.1-8B-Instant, Llama-3.3-70B-Versatile, Llama-4-Scout-17B-16E-Instruct, GPT-OSS-120B, and GPT-OSS-20B using a benchmark of 9,000 evaluation samples comprising 1,000 seed questions and 8,000 perturbed variants under both Standard Prompting and Chain-of-Thought (CoT) Prompting. Experimental results show that GPT-OSS-20B achieves the highest seed question accuracy of 96.08% under Standard Prompting, while larger models such as Llama-3.3-70B and GPT-OSS-120B demonstrate superior robustness across perturbation types. Furthermore, CoT prompting substantially improves reasoning for most models compared to Standard Prompting, yet a notable performance gap persists across all models relative to their English benchmarks, underscoring the inherent difficulty of perturbed Bengali text. This research makes a foundational contribution by providing GSM-PLUS-BN as a new resource and baseline for future Bengali mathematical reasoning research.
Chinese Translation
大型语言模型（LLMs）中数学推理的评估主要集中在英语等高资源语言上。这为在语言多样性丰富的地区（如孟加拉国，超过2.3亿人讲孟加拉语）公平地发展和部署人工智能创造了重大障碍。尽管这一全球性问题重要，但在孟加拉语数学推理方面的先前研究极为有限，且尚无系统性基准测试的扰动孟加拉数学数据集，这在评估模型的鲁棒性和真实理解能力（超越模式识别）方面留下了关键空白。本研究通过引入GSM-Plus-BN，填补了这一空白。GSM-Plus-BN是一个新颖的扰动孟加拉数学数据集，源自英语GSM-Plus基准，并经过人工翻译验证。我们使用包含1,000个种子问题和8,000个扰动变体的9,000个评估样本，对六个开源LLMs（Qwen3-32B、Llama-3.1-8B-Instant、Llama-3.3-70B-Versatile、Llama-4-Scout-17B-16E-Instruct、GPT-OSS-120B和GPT-OSS-20B）进行评估，采用标准提示和思维链（CoT）提示两种方式。实验结果表明，GPT-OSS-20B在标准提示下实现了96.08%的最高种子问题准确率，而像Llama-3.3-70B和GPT-OSS-120B等更大模型在各种扰动类型中表现出更好的鲁棒性。此外，与标准提示相比，CoT提示显著提高了大多数模型的推理能力，但所有模型相对于其英语基准仍存在显著的性能差距，突显了扰动孟加拉文本的固有难度。本研究通过提供GSM-PLUS-BN作为未来孟加拉数学推理研究的新资源和基准，做出了基础性贡献。
cs.CL / 10 / 2607.13260
Discourse-Aware Policy Analysis with Argumentation: A Hybrid LLM-Symbolic Framework for Disaster Governance
基于话语的政策分析与论证：灾害治理的混合LLM-符号框架
Abstract
Policy documents shape governance outcomes, but their reasoning is often implicit. Participatory commitments and managerial control routinely coexist in the same text, and the tensions between them are rarely stated directly. Existing computational approaches to policy discourse cannot express the frame-mediated relations that drive these tensions, where one argument narrows or instrumentalizes another rather than rejecting it. End-to-end summarization by large language models produces fluent text but offers little structure that domain experts can inspect or contest. We present Apaf, a hybrid LLM--symbolic pipeline that operationalizes critical discourse analysis as a quantitative bipolar argumentation framework over policy text. Arguments are first classified into deliberative or managerial frames. Four frame-mediated relation subtypes (agency reduction, agenda shift, instrumental support, and normative support) are then produced by deterministic rules over LLM-extracted features. We release a novel dataset of 100 sub-documents of disaster-risk-reduction policy from the USA, UK, Canada, and Australia, and show that the resulting argument graphs are accurate, interpretable, and stable across jurisdictions.
Chinese Translation
政策文件塑造治理结果，但其推理往往是隐含的。参与性承诺和管理控制通常在同一文本中共存，而它们之间的紧张关系很少被直接表述。现有的政策话语计算方法无法表达驱动这些紧张关系的框架中介关系，其中一个论点缩小或工具化另一个论点，而不是直接拒绝它。大型语言模型的端到端摘要生成流畅，但提供的结构很少，领域专家难以检查或质疑。我们提出了Apaf，一个混合的LLM-符号管道，将批判性话语分析操作化为政策文本上的定量双极论证框架。论点首先被分类为审议性或管理性框架。然后，通过对LLM提取特征的确定性规则生成四种框架中介关系子类型（代理减少、议程转移、工具性支持和规范性支持）。我们发布了一份来自美国、英国、加拿大和澳大利亚的100份灾害风险减少政策的子文档的新数据集，并展示了生成的论证图在各个司法管辖区内是准确的、可解释的和稳定的。
cs.CL / 11 / 2607.13311
Finding the Right Tables and Columns: A Benchmark and Corpus-Adaptive Embeddings for SQL Schema Retrieval
寻找合适的表和列：用于 SQL 架构检索的基准和语料库自适应嵌入
Abstract
Retrieval in the SQL setting has largely been studied as the task of finding, within a large collection of SQL statements, the statement that answers a natural-language question. At scale, however, a more fundamental retrieval problem precedes generation: schema retrieval, identifying the tables and columns a question requires in a database that may contain thousands of them, far more than fit in a model's context. We argue that this step warrants first-class evaluation. To this end, we recast five text-to-SQL datasets (Spider, BIRD, BEAVER, and two LiveSQLBench variants) as retrieval tasks at both table and column granularity, covering realistic and enterprise-scale schemas under two document representations, and we show that off-the-shelf text and code embedders transfer poorly to this setting. We then propose corpus-adaptive fine-tuning: natural-language queries are synthesized directly from the target schema corpus, granularity-aware hard negatives are mined, and a 305M-parameter embedder is fine-tuned contrastively. This procedure raises average recall@10 from 60.4 to 75.6 (nDCG@10 from 51.9 to 68.0), making the 305M model the strongest retriever under one billion parameters and competitive with state-of-the-art embedders of 4-8B parameters, more than an order of magnitude larger. The same recipe improves an 8B state-of-the-art embedder from 77.8 to 78.4 recall@10, matching the best result on the benchmark and indicating that the adaptation is backbone-agnostic. Leave-one-corpus-out experiments and a leakage audit show that these gains reflect a transferable schema-retrieval ability rather than memorization of the evaluation data. Our results establish schema linking as a standalone retrieval task and lightweight, label-free corpus adaptation as a practical route to deploying it at enterprise scale.
Chinese Translation
在 SQL 环境中的检索主要被研究为在大量 SQL 语句中找到能够回答自然语言问题的语句。然而，在大规模的情况下，更为根本的检索问题是架构检索，即识别一个问题在可能包含成千上万张表和列的数据库中所需的表和列，这远远超出了模型的上下文容量。我们认为这一步骤值得进行一流的评估。为此，我们将五个文本到 SQL 数据集（Spider、BIRD、BEAVER 以及两个 LiveSQLBench 变体）重新构建为表和列粒度的检索任务，涵盖现实和企业级架构，并在两种文档表示下进行实验，结果表明现成的文本和代码嵌入器在这一设置下表现不佳。随后，我们提出了语料库自适应微调：自然语言查询直接从目标架构语料库中合成，挖掘粒度感知的困难负样本，并对一个305M参数的嵌入器进行对比微调。这一过程将平均 recall@10 从 60.4 提升至 75.6（nDCG@10 从 51.9 提升至 68.0），使得305M模型成为在十亿参数下最强的检索器，并与4-8B参数的最先进嵌入器竞争，后者的规模大于一个数量级。相同的方法将一个8B的最先进嵌入器的 recall@10 从 77.8 提升至 78.4，匹配基准测试中的最佳结果，并表明这种适应性与基础架构无关。留一语料库实验和泄漏审计表明，这些提升反映了可转移的架构检索能力，而非对评估数据的记忆。我们的结果确立了架构链接作为独立的检索任务，并将轻量级、无标签的语料库适应性作为在企业规模上部署的实用途径。
cs.CL / 12 / 2607.13315
Meta-Learning Preferences for Multilingual LLM Alignment
多语言大语言模型对齐的元学习偏好
Abstract
Unequal availability of human preference data across languages poses a significant challenge for aligning large language models in multilingual settings. To address the lack of sufficient data in low-resource language alignment, we propose a meta-learning framework for Reinforcement Learning from Human Feedback and Direct Preference Optimization. By leveraging preference data from other languages, our framework learns a transferable initialization that enables effective adaptation to a target language with minimal data. We provide theoretical guarantees for both the meta-reward modeling and meta-policy optimization settings, and empirically demonstrate the effectiveness of our approach on multilingual benchmarks. In an extremely low-resource setting with only 100 target-language preference samples, our approach achieves up to $28\%$ win-rate improvements over baseline methods, and consistently outperforms baselines across multiple target languages and model scales. Our approaches retain these advantages across different combinations of meta-training languages and varying linguistic distances from the target languages.
Chinese Translation
人类偏好数据在不同语言中的不平衡可用性对多语言环境中大型语言模型的对齐构成了重大挑战。为了解决低资源语言对齐中数据不足的问题，我们提出了一种用于人类反馈强化学习和直接偏好优化的元学习框架。通过利用其他语言的偏好数据，我们的框架学习到了一种可转移的初始化，使其能够在数据极少的情况下有效适应目标语言。我们为元奖励建模和元策略优化设置提供了理论保证，并在多语言基准测试中实证展示了我们方法的有效性。在仅有100个目标语言偏好样本的极低资源环境中，我们的方法在基线方法上实现了高达28%的胜率提升，并在多个目标语言和模型规模上持续优于基线。我们的方案在不同的元训练语言组合和与目标语言的语言距离变化中保持了这些优势。
cs.CL / 13 / 2607.13347
Evaluation Ability Does Not Imply Optimization Utility: LLM-as-a-Judge Signals in Closed-Loop Table Recognition
评估能力并不意味着优化效用：闭环表格识别中的 LLM 作为裁判的信号
Abstract
LLM-as-a-judge is widely used to provide feedback and selection signals in closedloop regeneration, but this use remains insufficiently validated. We study it in table recognition, where deterministic TEDS evaluation provides a controlled testbed, using FinTabNet and OmniDocBench. Three findings emerge. First, judge signals were weak on both datasets: scores frequently tied, rankings were not reproducible, and the only selection policy that beat random on both datasets depended on an earliest-iteration tie rule, so its advantage cannot be attributed to the judge scores alone. Iteration produced better candidates, but the judge failed to recover them. Second, severe losses occurred even without specific judge feedback. A structurepreserving instruction significantly reduced the severe-loss rate on FinTabNet and was directionally consistent on OmniDocBench. The contrasts support target-preservation failure under unconstrained regeneration as a proximate mechanism of the observed severe losses. Third, the structure-preservation constraint reduced the severe-loss tail but produced no improvement. In an exploratory 2x2 analysis, the same protection was not stably observed when judge feedback was retained. These results do not dispute the value of LLMs as evaluators. Instead, they show that evaluation ability does not imply optimization utility. Iterative refinement requires, at minimum, a verification signal that deterministically detects structural change, rather than judge scores alone.
Chinese Translation
LLM 作为裁判被广泛用于在闭环再生中提供反馈和选择信号，但这种用法仍然缺乏充分验证。我们在表格识别中研究这一点，其中确定性的 TEDS 评估提供了一个受控的测试平台，使用 FinTabNet 和 OmniDocBench。研究得出了三个发现。首先，在两个数据集上裁判信号较弱：得分经常平局，排名不可重复，唯一在两个数据集上超越随机选择的选择策略依赖于最早迭代的平局规则，因此其优势不能仅归因于裁判得分。迭代产生了更好的候选项，但裁判未能恢复它们。第二，即使没有特定的裁判反馈，严重损失也会发生。一个保持结构的指令显著降低了 FinTabNet 上的严重损失率，并在 OmniDocBench 上方向一致。对比结果支持在无约束再生下目标保持失败作为观察到的严重损失的近因机制。第三，结构保持约束减少了严重损失的尾部，但没有产生改善。在一个探索性的 2x2 分析中，当保留裁判反馈时，同样的保护并未稳定观察到。这些结果并不否定 LLM 作为评估者的价值。相反，它们表明评估能力并不意味着优化效用。迭代改进至少需要一个验证信号，该信号能够确定性地检测结构变化，而不仅仅依赖裁判得分。
cs.CL / 14 / 2607.13372
A POS Tier Is the Key to Automated Annotation for Low-Resource Language Documentation: Neural Interlinear Glossing for Irabu, a Southern Ryukyuan Language
POS层是低资源语言文献自动注释的关键：针对南琉球语言Irabu的神经逐行注释
Abstract
Discourse data are the primary empirical basis of grammar writing in field linguistics, but producing interlinearized text is notoriously expensive - on the order of one hour of work per minute of recording. For endangered languages, where the time remaining to verify analyses with native speakers is itself limited, automating parts of the interlinearization workflow has direct documentary value. We implement a full neural annotation pipeline (morpheme segmentation, POS tagging, glossing) for Irabu Ryukyuan using deliberately small, transparent BiLSTM-CRF models, and evaluate it under a realistic hard constraint: approximately one hour of fully annotated discourse as the entire supervised resource. Two factors of the annotation itself are manipulated: its richness (with or without a POS tier) and its quantity (training budgets from 6 to 47 minutes). Gold POS improves grammatical glossing by +4.4 (SD 0.7) points (significant in all 5 seeds), and the gain grows as data shrink (+11.6 points at a quarter of the data); a POS tier more than halves the amount of glossed data needed to reach a given accuracy. In a fully automatic pipeline this gain is not yet realized: the tagger still errs on 12% of morphemes, and an incorrect POS misleads the glossing model more than no POS at all. The value is latent rather than lost: degrading gold POS with controlled noise shows the gain returning as tagger accuracy rises, with break-even near our tagger's current 88% and +1.6 to +3.2 points recovered at 92-96%. We conclude with a concrete recommendation for documentation practice: annotate quadrilinearly - text, POS, gloss, translation.
Chinese Translation
话语数据是田野语言学中语法撰写的主要实证基础，但生成逐行文本的成本极高——每分钟录音大约需要一个小时的工作时间。对于濒危语言而言，验证与母语者的分析所需的时间本身是有限的，因此自动化逐行注释工作流程的部分环节具有直接的文献价值。我们为Irabu琉球语实现了一个完整的神经注释管道（词素分割、词性标注、注释），使用故意设计的小型透明BiLSTM-CRF模型，并在一个现实的严格约束下进行评估：大约一小时的完全注释话语作为整个监督资源。我们操控了注释本身的两个因素：其丰富性（有或没有POS层）和数量（训练预算从6分钟到47分钟）。金标准词性标注（Gold POS）使语法注释提高了+4.4（标准差0.7）分（在所有5个种子中均显著），而随着数据量缩小，增益进一步增加（在四分之一数据时提高了+11.6分）；POS层使达到给定准确度所需的注释数据量减少了一半以上。在完全自动化的管道中，这一增益尚未实现：标注器在12%的词素上仍然出错，错误的词性标注比没有词性标注更容易误导注释模型。该价值是潜在的而非丧失的：通过控制噪声降低金标准词性标注显示，随着标注器准确性的提高，增益会恢复，在我们标注器当前的88%时接近平衡，92-96%时恢复+1.6到+3.2分。我们以对文献实践的具体建议作为结论：进行四线注释——文本、词性、注释、翻译。
cs.CL / 15 / 2607.13394
GFlowRL: Scaling Distribution-Matching RL to Large Language Models
GFlowRL：将分布匹配强化学习扩展到大型语言模型
Abstract
Generative Flow Networks (GFlowNets) offer a promising alternative to reward-maximizing reinforcement learning (RL) for large reasoning models, encouraging diverse reasoning paths by matching reward distributions rather than collapsing to dominant modes. Recent work shows promise on math and code, but scaling GFlowNet-style RL to modern post-training pipelines remains difficult: as model size, rollout horizon, reward noise, and distributed-systems complexity grow together, a learned prompt-conditional partition function becomes a source of gradient instability and engineering overhead rather than a useful normalizer. Through systematic analysis, we find that the learned partition function, previously treated as essential, can be replaced by an in-batch Monte Carlo estimate computed from the rollout group already required for training. We propose GFlowRL, a streamlined GFlowNet-style RL algorithm that removes the auxiliary partition network entirely while preserving the reward-distribution-matching objective, completed by two stabilizers: importance-sampling correction for rollout/trainer drift and asymmetric flow-gap clipping for outlier residuals. GFlowRL exceeds all counterparts on math, code, and adversarial red-teaming benchmarks, reaching a Codeforces rating of 2048 at the 14B scale (within 25 Elo of o3-mini) and attaining the highest average ASR@1 on AdvBench and HarmBench, outperforming the previous SOTA multi-turn attacker in a regime where FlowRL, a prior GFlowNet-style method, diverges. The same recipe transfers to all evaluated MoE configurations up to 235B parameters, where FlowRL again fails to converge. To our knowledge, GFlowRL is the first GFlowNet-style RL algorithm to scale stably across both dense and sparse architectures. Code will be at: https://github.com/microsoft/gflowrl
Chinese Translation
生成流网络（GFlowNets）为大型推理模型提供了一种有前景的替代方案，鼓励通过匹配奖励分布而非陷入主导模式来实现多样化的推理路径。近期的研究在数学和代码方面显示出良好的前景，但将GFlowNet风格的强化学习扩展到现代后训练管道仍然困难：随着模型规模、回滚时间、奖励噪声和分布式系统复杂性的共同增长，学习的条件分区函数成为梯度不稳定和工程开销的来源，而不是一个有用的归一化器。通过系统分析，我们发现先前被视为必要的学习分区函数可以被从已经用于训练的回滚组计算的批内蒙特卡洛估计所替代。我们提出了GFlowRL，这是一种简化的GFlowNet风格的强化学习算法，完全去除了辅助分区网络，同时保留了奖励分布匹配目标，并通过两个稳定器完成：用于回滚/训练者漂移的重要性采样修正和用于异常残差的非对称流差裁剪。GFlowRL在数学、代码和对抗性红队基准测试中超越了所有对手，在14B规模上达到了2048的Codeforces评级（比o3-mini高出25 Elo），并在AdvBench和HarmBench上获得了最高的平均ASR@1，超越了在FlowRL（之前的GFlowNet风格方法）发散的情况下的SOTA多轮攻击者。相同的方法适用于所有评估的MoE配置，参数规模高达235B，其中FlowRL再次未能收敛。据我们所知，GFlowRL是第一个在稠密和稀疏架构中稳定扩展的GFlowNet风格的强化学习算法。代码将在：https://github.com/microsoft/gflowrl
cs.CL / 16 / 2607.13399
Demystifying On-Policy Distillation: Roles, Pathologies, and Regulations
揭示在线策略蒸馏：角色、病理与调节
Abstract
On-policy distillation (OPD) has become a key paradigm in LLM post-training, yet its training dynamics remain poorly understood. We present a systematic study examining the role, pathologies, and regulations of OPD. We first clarify the role of OPD as an exploration catalyst: it steers the student toward correct reasoning paths via dense token-level guidance, without expanding capability ceiling. We confirm this by showing that prompt diversity matters more than per-problem sampling numbers, and critically, that the effectiveness of OPD hinges entirely on the quality of its guiding signal. This dependency exposes two pathologies that derail exploration. The Student-Teacher Mismatch occurs when a large teacher-student distributional gap causes the guiding signal to misalign with task correctness, steering exploration in counterproductive directions. Length Exploitation arises when the aggregated token-level objective creates length-dependent shortcuts, allowing the student to game the reward landscape through response truncation or redundant padding, exploring degenerate length modes rather than reasoning strategies. To tame these pathologies, we investigate lightweight signal regulations: advantage clipping and log-scale compression, ensuring exploration is guided by faithful signals. Experiments across seven benchmarks demonstrate that these regulations alleviate length exploitation and enable effective distillation, stably surpassing OPD variants and RLVR baselines, thereby confirming that well-regulated signal quality, rather than mere teacher scale, governs successful exploration in OPD.
Chinese Translation
在线策略蒸馏（On-policy distillation, OPD）已成为大规模语言模型（LLM）后训练中的关键范式，但其训练动态仍然不甚明了。我们进行了一项系统研究，考察了OPD的角色、病理和调节。我们首先澄清了OPD作为探索催化剂的角色：它通过密集的标记级指导，引导学生朝向正确的推理路径，而不扩展能力的上限。我们通过展示提示多样性比每个问题的采样数量更为重要，确认了这一点，并且关键的是，OPD的有效性完全依赖于其指导信号的质量。这种依赖性暴露了两种妨碍探索的病理。学生-教师不匹配（Student-Teacher Mismatch）发生在当教师与学生之间存在较大的分布差距时，导致指导信号与任务正确性不一致，从而引导探索朝着无益的方向。长度利用（Length Exploitation）则是在聚合的标记级目标创建了依赖长度的捷径，使得学生能够通过响应截断或冗余填充来操控奖励景观，探索退化的长度模式而非推理策略。为了驯服这些病理，我们研究了轻量级信号调节：优势裁剪（advantage clipping）和对数尺度压缩（log-scale compression），确保探索受到真实信号的引导。七个基准测试的实验表明，这些调节缓解了长度利用，并实现了有效的蒸馏，稳定地超越了OPD变体和RLVR基线，从而确认了良好调节的信号质量，而非单纯的教师规模，主导了OPD中的成功探索。
cs.CL / 17 / 2607.13430
Exploring Post-Training Alignment of Small Language Models for Biomedical Data-to-Text Generation: A Case Study of Medication Leaflet
探索小型语言模型在生物医学数据到文本生成中的后训练对齐：以药品说明书为例
Abstract
Translating complex biomedical data into patient-friendly narratives is central to modern biomedical informatics. This study presents a comparative analysis of training small language models (SLMs) in specialized biomedical datato-text generation tasks. We explore widely adopted post-training methods including supervised fine-tuning (SFT), direct preference optimization (DPO), odds ratio preference optimization (ORPO), and group relative policy optimization (GRPO) with Qwen-based SLMs on a medicine package leaflets dataset. To assess cross-dataset generalizability, we also curated drug label data from openFDA. We evaluate models using both standard lexical overlap metrics like ROUGE as well as semantic similarity measures. Across our experiments, the results show that (1) the aligned SLMs outperform proprietary models like GPT-5; (2) ORPO outperforms the SFTbaselines; (3) GRPO yields the most robust cross-dataset performance among the alignment methods tested as well as GPT-5.
Chinese Translation
将复杂的生物医学数据转化为患者友好的叙述是现代生物医学信息学的核心。本研究对在专业生物医学数据到文本生成任务中训练小型语言模型（SLMs）进行了比较分析。我们探讨了广泛采用的后训练方法，包括监督微调（SFT）、直接偏好优化（DPO）、优势比偏好优化（ORPO）和组相对策略优化（GRPO），并在药品包装说明书数据集上应用基于Qwen的SLMs。为了评估跨数据集的泛化能力，我们还从openFDA整理了药物标签数据。我们使用标准的词汇重叠指标（如ROUGE）以及语义相似性度量来评估模型。在我们的实验中，结果表明：（1）对齐的SLMs优于像GPT-5这样的专有模型；（2）ORPO优于SFT基线；（3）在测试的对齐方法中，GRPO在跨数据集性能方面表现最为稳健，超越了GPT-5。
cs.CL / 18 / 2607.13433
When Rubrics Change: Cross-Rubric Generalization for Critical Thinking Essay Scoring
当评分标准变化时：批判性思维论文评分的跨评分标准泛化
Abstract
Automated essay scoring (AES) research has largely focused on cross-prompt generalization, where essays from unseen prompts are scored while the scoring criteria are typically held constant. In practice, however, educators may revise or even introduce new rubrics in their scoring task, to evaluate different aspects of essays. We study cross-rubric generalization: training on essays labeled under one set of rubrics and evaluating on previously unseen rubrics, which target different aspects of the essay. We use a Large Language Model (LLM) fine-tuning framework with two components: rubric-agnostic intermediate representations, called traits, and target-essay supervision under seen rubrics during training. On an AES dataset augmented with multiple rubric-defined labels of student critical thinking skills, we find that traits improve macro F1 by 5.0% over a baseline without traits in the hardest setting, where both target rubrics and target essays are unseen during training. We further find that increasing target-essay supervision improves performance, with our best fine-tuned open-source Llama-based model outperforming GPT-5-mini prompting by 2.1% macro F1 and trailing GPT-5 by 1.9%. These results show that trait-based intermediate structure and controlled supervision improve generalization to unseen rubrics.
Chinese Translation
自动化论文评分（AES）研究主要集中在跨提示泛化，即对未见提示的论文进行评分，而评分标准通常保持不变。然而，在实际操作中，教育工作者可能会修改或甚至引入新的评分标准，以评估论文的不同方面。我们研究跨评分标准泛化：在一组评分标准下对标记的论文进行训练，并在先前未见的评分标准下进行评估，这些评分标准针对论文的不同方面。我们使用一个大型语言模型（LLM）微调框架，其中包含两个组件：与评分标准无关的中间表示，称为特征，以及在训练期间根据已见评分标准的目标论文监督。在一个增强了多种评分标准定义的学生批判性思维技能标签的AES数据集上，我们发现特征在最困难的设置中（在该设置中，目标评分标准和目标论文在训练期间均未见）相比于没有特征的基线提高了5.0%的宏F1分数。我们进一步发现，增加目标论文的监督可以提高性能，我们最佳的微调开源Llama模型在宏F1分数上比GPT-5-mini提示高出2.1%，而与GPT-5相比低了1.9%。这些结果表明，基于特征的中间结构和受控监督改善了对未见评分标准的泛化。
cs.CL / 19 / 2607.13457
Live Gurbani Tracking: A Benchmark and Reference System for Captioning Sikh Kirtan
实时古尔巴尼跟踪：锡克教基尔坦字幕的基准和参考系统
Abstract
We present a benchmark and reference system for live captioning of Sikh Kirtan - the continuous, sung recitation of verses from the Sri Guru Granth Sahib Ji (SGGS). Unlike open-vocabulary lyrics transcription, Kirtan captioning is a closed-vocabulary problem: every displayed line must be an exact, word-for-word line from the canonical scripture, because displaying misspelled Gurmukhi is considered religiously inappropriate. We formalize the task as predicting, at every time t, a pair (shabad_id, line_idx) or null, and organize the problem space into a 2x2 matrix along two orthogonal axes: live vs. offline (causal vs. full-audio access) and blind vs. oracle (shabad identity discovered vs. given). We release v1 of the benchmark - 4 hand-annotated Kirtan recordings x 3 cold-start offsets = 12 evaluation cases, ~57 minutes of scored audio - together with a scorer that computes frame accuracy at 1s resolution over a scored region, with a 1s collar and gap-tolerant scoring at segment boundaries. We describe a reference system (fine-tuned 120M IndicConformer -> fuzzy matcher -> state machine; INT8 ONNX; RTF ~0.05 on one Apple Silicon core) that achieves 57.9% overall frame accuracy across all 12 cases (10/12 correct shabad locks) on the hardest variant (live x blind). We compare against three trivial baselines (empty, shifted-5s, perfect) and discuss why standard ASR metrics (WER/CER) measure transcription accuracy rather than the display accuracy this task requires. The benchmark, reference system, and a live deployment are released under permissive licenses to facilitate further improvements.
Chinese Translation
我们提出了一个用于锡克教基尔坦（Kirtan）实时字幕的基准和参考系统——这是对《斯里古鲁格兰特·萨希布》（Sri Guru Granth Sahib Ji, SGGS）经文的连续吟唱。与开放词汇歌词转录不同，基尔坦字幕是一个封闭词汇问题：每一行显示的内容必须是经典经文中的确切逐字行，因为显示拼写错误的古尔穆基（Gurmukhi）被视为宗教不当。我们将该任务形式化为在每个时间点 t 预测一对 (shabad_id, line_idx) 或 null，并将问题空间组织成一个沿两个正交轴的 2x2 矩阵：实时 vs. 离线（因果 vs. 完整音频访问）和盲 vs. 预言（shabad 身份发现 vs. 给定）。我们发布了基准的 v1 版本——4 个手动注释的基尔坦录音 x 3 个冷启动偏移 = 12 个评估案例，约 57 分钟的评分音频——以及一个评分器，该评分器在评分区域内以 1 秒分辨率计算帧准确率，具有 1 秒的边界容忍评分和间隙容忍评分。我们描述了一个参考系统（微调的 120M IndicConformer -> 模糊匹配器 -> 状态机；INT8 ONNX；在一个 Apple Silicon 核心上的 RTF ~0.05），在最难的变体（实时 x 盲）中在所有 12 个案例中实现了 57.9% 的整体帧准确率（10/12 个正确的 shabad 锁）。我们与三个简单的基线（空、偏移 5 秒、完美）进行了比较，并讨论了为什么标准的自动语音识别（ASR）指标（WER/CER）测量的是转录准确性，而不是该任务所需的显示准确性。基准、参考系统和实时部署在宽松许可证下发布，以促进进一步的改进。
cs.CL / 20 / 2607.13465
DevicesWorld: Benchmarking Cross-Device Agents in Heterogeneous Environments
DevicesWorld：异构环境中跨设备智能体的基准测试
Abstract
LLM-based agents have rapidly improved at operating individual digital environments such as mobile applications, desktop systems, and smart homes. However, real-world user goals often span multiple devices: information may come from a phone, be processed on a desktop, and the result may need to appear on another device. Most existing benchmarks center on a single dominant execution environment, making it difficult to evaluate whether agents can acquire and integrate information across heterogeneous devices and complete end-to-end tasks with cross-device dependencies. We introduce DevicesWorld, a large-scale executable benchmark for cross-device collaborative operation. DevicesWorld contains 6,140 tasks and integrates three classes of device environments -- mobile, desktop, and IoT -- into a unified cross-device interaction and evaluation framework. Each task defines a natural-language user goal, participating devices and initial states, executable actions, rule-based verifiers, and a cleanup procedure. A multi-stage construction and quality-control pipeline keeps tasks close to realistic user needs while allowing final outcomes to be automatically verified from device states and generated files. We evaluate five frontier LLM-agent systems on a fixed evaluation set. All methods achieve low success rates, with the best reaching only 12.5%. Among failed runs, about 28.7% satisfy at least one scoring condition yet still fail the full task. Trajectories show that agents become stuck acquiring information or manipulating interfaces, confuse source and output devices, or terminate before all conditions are jointly satisfied. DevicesWorld turns cross-device collaborative operation into an executable, reproducible, and diagnostically useful evaluation problem for research on reliable cross-device agents.
Chinese Translation
基于大规模语言模型（LLM）的智能体在操作单一数字环境（如移动应用、桌面系统和智能家居）方面迅速取得了进展。然而，现实世界中的用户目标往往跨越多个设备：信息可能来自手机，在桌面上处理，结果可能需要在另一个设备上显示。现有的大多数基准测试集中于单一主导执行环境，这使得评估智能体是否能够跨异构设备获取和整合信息，以及完成具有跨设备依赖性的端到端任务变得困难。我们提出了DevicesWorld，这是一个用于跨设备协作操作的大规模可执行基准测试。DevicesWorld包含6140个任务，并将移动、桌面和物联网（IoT）三类设备环境整合到一个统一的跨设备交互和评估框架中。每个任务定义了一个自然语言用户目标、参与设备和初始状态、可执行动作、基于规则的验证器以及清理程序。一个多阶段的构建和质量控制流程使得任务与现实用户需求保持接近，同时允许从设备状态和生成文件中自动验证最终结果。我们在固定评估集上评估了五个前沿的LLM智能体系统。所有方法的成功率都很低，最佳方法仅达到12.5%。在失败的运行中，大约28.7%满足至少一个评分条件但仍然未能完成整个任务。轨迹显示，智能体在获取信息或操控界面时陷入困境，混淆了源设备和输出设备，或在所有条件未共同满足之前就终止。DevicesWorld将跨设备协作操作转变为一个可执行、可重复且具有诊断价值的评估问题，为可靠的跨设备智能体研究提供支持。
cs.CL / 21 / 2607.13474
MyAG: A Graph-Based Framework for Designing and Analyzing Composable LLM Agent Systems
MyAG：一个基于图的可组合 LLM 代理系统设计与分析框架
Abstract
We present MyAG, a graph-based framework for designing and analyzing composable LLM agent systems. Our framework separates agent system construction into three graph abstractions: a component graph for agents, environments, and modules; a workflow graph for execution control; and a search graph for runtime execution. This separation allows users to flexibly reuse the same components with different strategies. We further support hierarchical composition through recursive system nodes and provide monitoring and visualization tools for inspecting agent execution. Experiments on representative agent applications show that our framework supports flexible agent system design and helps analyze performance-efficiency tradeoffs. Our framework is publicly available and fully open-source.
Chinese Translation
我们提出了 MyAG，一个基于图的可组合 LLM 代理系统设计与分析框架。我们的框架将代理系统的构建分为三个图抽象：用于代理、环境和模块的组件图；用于执行控制的工作流图；以及用于运行时执行的搜索图。这种分离使用户能够灵活地以不同策略重用相同的组件。我们进一步通过递归系统节点支持层次化组合，并提供监控和可视化工具以检查代理执行。对代表性代理应用的实验表明，我们的框架支持灵活的代理系统设计，并有助于分析性能与效率之间的权衡。我们的框架是公开可用的，并且完全开源。
cs.CL / 22 / 2607.13551
Cost-Pragmatic Quality Gating and Selection-Fusion Multi-Model Combiners for BioASQ Phases A+ and B
成本务实的质量门控与选择-融合多模型组合器用于BioASQ阶段A+和B
Abstract
We describe our BioASQ Task 14B 2026 system. The work centers on two design decisions: how aggressively to re-retrieve when first-stage retrieval is weak, and how to combine multiple language-model answers. Retrieval unions two parallel pipelines - a hybrid first stage (dense BGE + BM25 + RRF, reaching R@200 = 99.3% on the BioASQ-13b historical archive) and an agent-driven pipeline that decomposes the question over PubMed, Europe PMC, and iCite - with a BGE cross-encoder quality gate flagging weakly-supported questions for selective re-retrieval. On Task 12B 2024 validation, a cost-pragmatic re-retrieval policy beats a skill-strict baseline significantly on list F1 and list precision, at 12% lower re-retrieval cost. Holding prompt and model fixed across val and test 13B (different question sets), list F1 rises by +0.132 absolute on the BioASQ-released gold-input pool, consistent with substantial retrieval-side headroom. For Phase B answering we decompose multi-model ensemble lift into a selection component bounded by the per-question oracle and a fusion component that aggregators can exceed. The decomposition predicts before any experiment that LLM-as-judge wins on selection-dominated metrics (yes/no, multi-reference ROUGE) but is structurally insufficient on the recall component of fusion-friendly metrics (factoid rank-1, list recall). On Task 13B 2025 our synonym-union resolver wins list recall on every head, while GPT-5.5 solo retains the list-F1 lead because the resolver's wider item set costs precision. On the Task 14B 2026 preliminary leaderboard our team places first on the combined-exact aggregate on three of the eight (phase x batch) leaderboards, wins four individual question-type cells, and takes #1 on Phase B b3 ideal.
Chinese Translation
我们描述了我们的BioASQ任务14B 2026系统。该工作集中于两个设计决策：在第一阶段检索较弱时，如何积极地进行重新检索，以及如何结合多个语言模型的答案。检索结合了两个并行管道——一个混合的第一阶段（稠密BGE + BM25 + RRF，在BioASQ-13b历史档案上达到R@200 = 99.3%）和一个基于代理的管道，该管道在PubMed、Europe PMC和iCite上分解问题——并使用BGE交叉编码器质量门控标记支持较弱的问题以进行选择性重新检索。在任务12B 2024验证中，成本务实的重新检索策略在列表F1和列表精度上显著优于严格技能基线，且重新检索成本降低了12%。在保持提示和模型在验证和测试13B（不同问题集）中固定的情况下，列表F1在BioASQ发布的金输入池上绝对上升了+0.132，与显著的检索侧余量一致。对于阶段B回答，我们将多模型集成提升分解为一个由每个问题的oracle界定的选择组件和一个聚合器可以超越的融合组件。该分解在任何实验之前预测，LLM作为评判者在选择主导的指标（是/否，多参考ROUGE）上获胜，但在融合友好的指标的召回组件上结构上不足（事实排名1，列表召回）。在任务13B 2025中，我们的同义词联合解析器在每个头上都赢得了列表召回，而GPT-5.5单独保持了列表F1的领先，因为解析器更广泛的项目集降低了精度。在任务14B 2026的初步排行榜上，我们的团队在八个（阶段x批次）排行榜中的三个上获得了综合精确总分的第一名，赢得了四个单独问题类型单元，并在阶段B b3理想中获得了第一名。
cs.CL / 23 / 2607.13568
Graded Entity-Familiarity Readouts in Language Models: Polish Adaptation, Cross-Language Robustness, and Refusal Steering
语言模型中的分级实体熟悉度读取：波兰适应性、跨语言鲁棒性与拒绝引导
Abstract
Can a language model estimate its familiarity with an entity before generating an answer? We study activations at the final prompt token in twelve instruction-tuned models from the Bielik, PLLuM, Gemma-4, and Qwen3 families, using a new dataset of 1,440 Polish entities spanning four domains and ten Wikipedia-pageview deciles, plus fabricated controls. Familiarity-probe scores separate real from fabricated entities in every family; in the Polish-adapted Bielik and PLLuM families they additionally track entity popularity (model-mean Spearman $\rho$ 0.28-0.57, versus at most 0.11 in Gemma-4 and Qwen3), a pattern more strongly associated with Polish adaptation than with parameter count in this model sample. In a paired experiment on two families, probes retain 96-101% of within-language AUROC when the Polish question stem is replaced with an English one around unchanged entity names, showing robustness to prompt language in this setting. In Gemma-4-12B, the only model that natively refuses, adding a one-dimensional familiarity direction at a single layer moves refusal rates monotonically in both directions (0.24 to 1.00 on well-known entities; 0.73 to 0.00 on unknown ones). Finally, a calibrated familiarity probe is competitive among pre-generation abstention gates, although post-generation detectors better predict behavioral error on average. These results support a graded pre-generation entity-familiarity readout, and a separation between representational familiarity and the policy that converts it into abstention.
Chinese Translation
语言模型能否在生成答案之前评估其对某个实体的熟悉度？我们研究了来自Bielik、PLLuM、Gemma-4和Qwen3四个模型家族的十二个指令调优模型在最后提示标记的激活情况，使用了一个包含1,440个波兰实体的新数据集，涵盖四个领域和十个维基百科页面浏览量的十分位，以及伪造的对照组。熟悉度探测分数在每个模型家族中都能区分真实和伪造的实体；在波兰适应的Bielik和PLLuM家族中，它们还跟踪实体的流行度（模型均值斯皮尔曼相关系数 $
ho$ 0.28-0.57，而Gemma-4和Qwen3最多为0.11），这种模式与波兰适应性比与该模型样本中的参数数量更强相关。在对两个家族的配对实验中，当波兰问题提示被替换为英语提示且实体名称不变时，探测器保持了96-101%的语言内AUROC，显示出在这种设置下对提示语言的鲁棒性。在Gemma-4-12B中，唯一一个本地拒绝的模型，在单层中添加一个一维熟悉度方向使得拒绝率在两个方向上单调变化（在知名实体上从0.24变为1.00；在未知实体上从0.73变为0.00）。最后，一个经过校准的熟悉度探测器在生成前的弃权门中具有竞争力，尽管生成后的检测器在平均上更好地预测行为错误。这些结果支持了分级生成前实体熟悉度的读取，并区分了表征熟悉度与将其转化为弃权的策略。
cs.CL / 24 / 2607.13591
Memory as a Controlled Process: Learned Adaptive Memory Management for LLM Agents
记忆作为一种受控过程：针对大型语言模型代理的自适应记忆管理
Abstract
Large Language Model (LLM) agents increasingly rely on external memory systems to accumulate experience across tasks. Yet nearly all existing approaches, from graph-structured memories to reflective insight stores, access memory through fixed, hand-designed heuristics. We argue that this static view of memory is a core bottleneck for agentic learning because optimal memory behavior is fundamentally context-dependent. The early stages of the tasks, benefit from minimal retrieval because memory is sparse; recurring goal types benefit from plan reuse rather than generic nearest-neighbor lookup; stuck agents benefit from re-retrieval with alternative queries; and across long task streams, the memory store itself must be consolidated and pruned to remain useful. We present Memory as a Controlled Process (MemCon), a framework that models memory operations as a Markov Decision Process and learns an online policy that adaptively decides when, what, and how much to retrieve, when to inject a distilled plan, and when to consolidate or forget. MemCon is backend-agnostic: it wraps any existing memory implementation, learns from task-by-task binary feedback with no pretraining and no additional LLM calls, and uses a lightweight tabular contextual bandit with UCB exploration that converges within tens of tasks. Across 6 benchmarks, 3 agent frameworks, and 3 LLM backbones, MemCon consistently outperforms multiple memory baselines by up to 15.2 points in task success while reducing token consumption by 5--20%.
Chinese Translation
大型语言模型（LLM）代理越来越依赖外部记忆系统来积累跨任务的经验。然而，几乎所有现有的方法，从图结构记忆到反思洞察存储，都是通过固定的、手工设计的启发式方法来访问记忆。我们认为，这种静态的记忆观是代理学习的一个核心瓶颈，因为最佳的记忆行为从根本上依赖于上下文。任务的早期阶段，由于记忆稀疏，最小的检索是有益的；重复的目标类型更依赖于计划重用，而不是通用的最近邻查找；被困的代理则受益于使用替代查询进行重新检索；在长任务流中，记忆存储本身必须进行整合和修剪以保持其有效性。我们提出了“记忆作为一种受控过程”（Memory as a Controlled Process, MemCon），这是一个将记忆操作建模为马尔可夫决策过程的框架，并学习一种在线策略，自适应地决定何时、检索什么以及检索多少，何时注入提炼的计划，以及何时进行整合或遗忘。MemCon与后端无关：它可以包装任何现有的记忆实现，利用逐任务的二元反馈进行学习，无需预训练和额外的LLM调用，并使用轻量级的表格上下文赌博机（tabular contextual bandit）结合UCB探索，在数十个任务内收敛。在6个基准测试、3个代理框架和3个LLM骨干网络中，MemCon在任务成功率上始终超越多个记忆基线，最高可达15.2个百分点，同时减少5%至20%的令牌消耗。
cs.CL / 25 / 2607.13602
Analogical Deep Research: Retrieving and Integrating Historical Analogies for Foresight Analysis
类比深度研究：检索与整合历史类比以进行前瞻性分析
Abstract
Systematic comparisons between current situations and structurally similar past events in the historical, i.e., historical analogies, is among the most powerful tools for foresight analysis. In this work, we present a new task called Analogical Deep Research (ADR) to Large Language Model (LLM) agents and construct the first ADR benchmark ADR-bench to study whether LLM agents are able to find and leverage historical analogies when doing foresight analysis. Our investigation reveals a key obstacle: LLM agents are poor at finding analogies because they match on surface features rather than underlying mechanisms. We argue that ADR is inherently a causal question as it requires understanding why the event occurred. Based on our theoretical analysis, we propose two principles required for ADR, including the mechanism alignment and cross-analogy confirmation. Built upon our theoretical results, we propose a new agentic framework called Causal Analogical Researcher (CANA) that guides LLMs to find and integrate historical analogies. CANA incorporates a simple yet effective structural decomposition representation, and integrates structural feedback for reflective improvements of historical analogy identification and integration. We show that CANA brings up to 10% improvements in historical analogy generation, and surpasses the state-of-the-art deep research agents in the ADR-bench. Case studies with the ongoing events confirm the effectiveness of CANA in leveraging historical analogies.
Chinese Translation
对当前情况与历史上结构相似的事件进行系统比较，即历史类比，是进行前瞻性分析的最强大工具之一。在本研究中，我们提出了一项新任务，称为类比深度研究（Analogical Deep Research, ADR），并构建了第一个ADR基准（ADR-bench），以研究大型语言模型（Large Language Model, LLM）代理在进行前瞻性分析时是否能够找到和利用历史类比。我们的调查揭示了一个关键障碍：LLM代理在寻找类比方面表现不佳，因为它们主要基于表面特征而非潜在机制进行匹配。我们认为，ADR本质上是一个因果问题，因为它需要理解事件发生的原因。基于我们的理论分析，我们提出了ADR所需的两个原则，包括机制对齐和跨类比确认。在我们的理论结果基础上，我们提出了一种新的代理框架，称为因果类比研究者（Causal Analogical Researcher, CANA），该框架指导LLM找到并整合历史类比。CANA结合了一种简单而有效的结构分解表示，并整合了结构反馈，以反思性地改善历史类比的识别和整合。我们展示了CANA在历史类比生成方面提高了多达10%的效果，并在ADR-bench中超越了最先进的深度研究代理。对正在进行的事件的案例研究确认了CANA在利用历史类比方面的有效性。
cs.CL / 26 / 2607.13683
Self-Evolving Agent Harnesses via Gated Semantic Quality-Diversity
自我进化代理的门控语义质量多样性利用
Abstract
An LLM agent's real-task performance is shaped as much by the harness around its model as by the frozen model itself: its prompts, injected knowledge, runtime control, and configuration. In deployment the harness is often the only lever available, so improving it automatically is the natural way to raise performance without touching the weights. The hard part is not generating changes but knowing which one truly helped. Self-generated feedback is noisy, and an apparent gain can be a measurement artifact or an edit that merely overfits the tasks it was tuned on. We present a self-evolving agent-harness framework that separates proposing changes from crediting them: a language model diagnoses failures and proposes patches, while all sampling, measurement, and significance testing are owned by deterministic code, so every credited improvement is trustworthy by construction. Patches populate a gated, categorical quality-diversity archive (GSME) keyed on the (WHERE x WHY) pathology an edit addresses rather than the tasks it fixes, an anti-overfitting inductive bias; generalization is measured on a sealed test scored only after evolution. Across seven domains with a frozen open-weight model, the harness is train-selected and scored once on a sealed test; its credited gains there are +9 to +15.5pp and retain 86-147% of the training gain, evidence they generalize rather than overfit. The winning patch tracks the model's dominant pathology, not its size or family: changing the model can change the pathology and the patch, while the same pathology-to-patch match recurs across two model families. What transfers is the diagnose-and-credit loop, not any specific harness.
Chinese Translation
大型语言模型（LLM）代理的实际任务表现不仅受其模型本身的影响，还受到围绕其模型的框架的影响：包括提示、注入的知识、运行时控制和配置。在部署中，框架往往是唯一可用的杠杆，因此自动改进它是提高性能的自然方式，而无需调整模型权重。困难之处在于，不仅要生成变化，还要知道哪种变化真正有帮助。自生成的反馈通常是噪声，表面上的增益可能是测量伪影或仅仅是过拟合于其调优任务的编辑。我们提出了一种自我进化代理框架，该框架将提出变化与归因分开：语言模型诊断失败并提出补丁，而所有采样、测量和显著性测试均由确定性代码负责，因此每个被认可的改进在构造上都是可信的。补丁填充在一个门控的分类质量多样性档案（GSME）中，该档案以编辑所针对的（WHERE x WHY）病理为键，而不是修复的任务，这是一种反过拟合的归纳偏差；泛化在一个密封测试上进行测量，该测试仅在进化后评分。在七个领域中，使用一个冻结的开放权重模型，框架经过训练选择并在一个密封测试上评分；其在该测试中的认可增益为+9到+15.5个百分点，并保留了86-147%的训练增益，证明它们是泛化而非过拟合。获胜的补丁跟踪模型的主要病理，而不是其大小或家族：改变模型可以改变病理和补丁，而相同的病理与补丁的匹配在两个模型家族中反复出现。转移的是诊断与归因的循环，而不是任何特定的框架。
cs.CL / 27 / 2607.13707
The Test Oracle Problem in Synthetic LLM-as-Judge Corpora: Disappearance, Distortion and a Validation Protocol
合成 LLM 作为评判者语料库中的测试oracle问题：消失、扭曲与验证协议
Abstract
Studies of bias in LLM-as-judge systems typically build synthetic corpora by prompting an LLM to generate a hallucinated answer to pair with a factual one, then presenting both to a judge. We report a case in which this generation step silently failed, and use it to argue that the failure mode is structural rather than incidental. In a multilingual (Turkish/English) faithfulness-judgment corpus, a decoding-budget parameter shared between judging and generation calls truncated one producer's hallucinated answers to a few words. The resulting items produced a large, statistically robust effect: a 32-point cross-lingual collapse in one judge's selection accuracy, replicated from N=50 to N=500, explained by a three-layer mechanistic account, and confirmed by a controlled producer-swap experiment, none of which was real. The effect vanished to ceiling once the shared parameter was corrected, and only manual reading of the raw generations, not any aggregate statistical check, exposed the fault. A second measured bias (Markdown-formatting preference) was not fabricated but distorted by the same fault, its magnitude and in one case its sign shifting with stimulus length, a mode aggregate metrics cannot distinguish from the first. We frame the underlying vulnerability using the test oracle problem: corpora whose negative examples are LLM-generated carry no mechanical way to verify item integrity, while corpora built by deterministic perturbation of a gold answer carry an item-level oracle for free. A positive control supports this claim directly: an analogous fault injected into a minimal perturbation-based corpus is caught with 100% accuracy by a zero-cost, zero-human gold-to-negative string comparison. We close with a validation protocol, derived from our own case, for analysts working in the oracle-less regime that we argue describes most contemporary multilingual LLM-as-judge corpora.
Chinese Translation
对 LLM 作为评判者系统中偏见的研究通常通过提示 LLM 生成虚构答案与事实答案配对，从而构建合成语料库，然后将两者呈现给评判者。我们报告了一个生成步骤静默失败的案例，并利用该案例论证失败模式是结构性的而非偶然的。在一个多语言（土耳其语/英语）忠实性判断语料库中，评判和生成调用之间共享的解码预算参数将一个生成者的虚构答案截断为几个词。结果产生了一个显著且统计上稳健的效应：在一个评判者的选择准确性上出现了32点的跨语言崩溃，从 N=50 复制到 N=500，解释为一个三层机制的解释，并通过一个受控的生产者交换实验确认，所有这些都不是真实的。当共享参数被修正后，该效应消失至上限，只有手动阅读原始生成内容，而不是任何聚合统计检查，才揭示了这个缺陷。第二个测量的偏见（Markdown 格式偏好）并非伪造，但同样受到该缺陷的扭曲，其幅度和在某些情况下其符号随着刺激长度的变化而变化，这种模式的聚合指标无法与第一个区分开。我们使用测试 oracle 问题框架来描述潜在的脆弱性：那些负例由 LLM 生成的语料库没有机械方式来验证项目的完整性，而通过对黄金答案的确定性扰动构建的语料库则免费携带项目级的 oracle。一个正向对照直接支持这一主张：注入到最小扰动基础语料库中的类似缺陷以 100% 的准确率被零成本、零人力的黄金与负字符串比较捕获。我们最后提出了一种验证协议，基于我们自己的案例，供在无 oracle 领域工作的分析师使用，我们认为这描述了大多数当代多语言 LLM 作为评判者语料库的情况。
cs.CL / 28 / 2607.13721
Self-supervised Speech Comparison for L2 Phone, Rhythm, and Intonation Scoring
自监督语音比较用于L2音位、节奏和语调评分
Abstract
L2 speech assessment has traditionally focused on phonetic assessment, leaving the scoring of suprasegmental features such as rhythm and intonation underexplored. Moreover, assessment methods often require training with labeled L2 speech data, making them difficult to apply in low-resource settings. We investigate whether DTW over self-supervised WavLM representations can provide a text-free framework for assessing phonetic accuracy, rhythm, and intonation in English and Japanese L2 speech. Results show that a basic DTW-based approach that compares learner speech to native templates exceeds human agreement on holistic and sentence-level phonetic scoring. For rhythm, we introduce methods that measure the degree of warping in the DTW alignment path; our best method approaches human-level performance. For intonation, we combine DTW distance over prosodic residuals with pitch and intensity features, but performance remains more modest on some tasks. Our results point to self-supervised representations as a promising, text-free basis for multi-aspect pronunciation assessment.
Chinese Translation
L2语音评估传统上侧重于音位评估，而对节奏和语调等超音段特征的评分则研究较少。此外，评估方法通常需要使用带标签的L2语音数据进行训练，这使得它们在资源匮乏的环境中难以应用。我们研究了基于自监督WavLM表示的动态时间规整（DTW）是否可以提供一种无文本框架，用于评估英语和日语L2语音的音位准确性、节奏和语调。结果表明，基本的DTW方法将学习者的语音与母语模板进行比较，其整体和句子级音位评分的结果超过了人类评估者的一致性。在节奏方面，我们引入了测量DTW对齐路径中扭曲程度的方法；我们最好的方法接近人类水平的表现。在语调方面，我们将基于韵律残差的DTW距离与音高和强度特征相结合，但在某些任务上的表现仍然较为温和。我们的结果表明，自监督表示作为多方面发音评估的有希望的无文本基础。
cs.CL / 29 / 2607.13753
Post-Training Shifts Confidence: A Three-Stage Analysis of How SFT, RL, and OPD Shape Pre-, Intra-, and Post-CoT Calibration
后训练信心的转变：SFT、RL和OPD如何塑造链式思维生成前、中、后的三阶段分析
Abstract
Large language models have made strong reasoning gains through supervised fine-tuning, reinforcement learning, and on-policy distillation, yet these post-training methods are usually evaluated only by final-answer accuracy. We study how they reshape confidence during reasoning. We introduce a three-stage calibration framework that evaluates confidence before, during, and after chain-of-thought generation, corresponding to difficulty estimation, early termination, and answer aggregation. Through a controlled comparison on mathematical reasoning benchmarks, we find that OPD provides the most useful pre-reasoning confidence, SFT gives the strongest online signal for early stopping, and RL produces the most reliable trace-level signal for aggregation. We further show that confidence reliability is position-dependent: RL confidence becomes informative after a path-commitment phase, while OPD confidence is useful early but can become inversely calibrated later. Based on this observation, we propose PosConf, a position-aware confidence strategy that uses confidence only from reliable relative-position intervals. PosConf improves RL answer aggregation by 6.1 points over majority voting and consistently improves OPD early stopping under tight token budgets, with gains up to 4.3 points by avoiding its later inverse-calibration region, showing that \emph{confidence in reasoning models should be used both stage-wise and position-awarely}. Our code is available at https://github.com/EIT-NLP/Post-Training-Calibration.
Chinese Translation
大型语言模型通过监督微调、强化学习和在线策略蒸馏在推理方面取得了显著进展，但这些后训练方法通常仅通过最终答案的准确性进行评估。我们研究了它们如何在推理过程中重塑信心。我们引入了一个三阶段的校准框架，评估在链式思维生成之前、期间和之后的信心，分别对应于难度估计、早期终止和答案聚合。通过对数学推理基准的控制比较，我们发现OPD提供了最有用的推理前信心，SFT为早期停止提供了最强的在线信号，而RL则产生了最可靠的逐步信号用于聚合。我们进一步表明，信心的可靠性依赖于位置：RL信心在路径承诺阶段后变得信息丰富，而OPD信心在早期有用，但在后期可能会变得反向校准。基于这一观察，我们提出了PosConf，一种位置感知的信心策略，仅使用来自可靠相对位置区间的信心。PosConf在多数投票基础上提高了RL答案聚合6.1个百分点，并在紧张的标记预算下持续改善OPD的早期停止，通过避免其后期反向校准区域，提升了最多4.3个百分点，表明“推理模型中的信心应同时在阶段上和位置上进行使用”。我们的代码可在 https://github.com/EIT-NLP/Post-Training-Calibration 获取。
cs.CL / 30 / 2607.13854
SPyCE: Skill-Policy Co-evolution for Multimodal Agents
SPyCE：多模态智能体的技能-策略共同进化
Abstract
Multimodal agents that think with images iteratively manipulate visual evidence and invoke tools across many steps. Existing reinforcement learning methods reduce trajectories to scalar rewards, forcing the policy to discover reusable tool-use patterns from scratch on every new task; memory-based alternatives retain past experience, yet they rely on test-time retrieval, without updating the policy to absorb reusable patterns from that experience. Our key insight is that multimodal reasoning trajectories should be distilled into reusable skills that co-evolve with the policy during training, rather than being consumed as rewards or retrieved from a static store. To this end, we propose SPyCE (Skill-Policy Co-evolution), a framework that distills trajectories into a hierarchical skill library and updates it throughout reinforcement learning. Execution skills capture local visual operations, while workflow skills encode high-level priors that orchestrate tool use. During training, the policy model conditions on retrieved skills to guide its rollouts, while the skill library evolves using valuable rollouts generated by the policy. This creates a closed loop in which improved policies yield better skills, and the evolving skill library, in turn, provides stronger priors for policy rollouts. Experiments across eight benchmarks demonstrate that SPyCE consistently outperforms both RL-based and memory-based baselines. Further analysis reveals that both the hierarchical skill design and the co-evolution mechanism are critical to our design. These results suggest joint skill-policy optimization as a promising paradigm for building capable multimodal agents.
Chinese Translation
多模态智能体通过图像进行思考，迭代地操控视觉证据并在多个步骤中调用工具。现有的强化学习方法将轨迹简化为标量奖励，迫使策略在每个新任务中从零开始发现可重用的工具使用模式；而基于记忆的替代方案则保留过去的经验，但它们依赖于测试时的检索，而不更新策略以吸收来自该经验的可重用模式。我们的关键见解是，多模态推理轨迹应被提炼为可重用的技能，这些技能在训练过程中与策略共同进化，而不是作为奖励被消耗或从静态存储中检索。为此，我们提出了SPyCE（技能-策略共同进化），一个将轨迹提炼为层次化技能库并在整个强化学习过程中更新的框架。执行技能捕捉局部视觉操作，而工作流技能编码高层先验，以协调工具使用。在训练过程中，策略模型依赖于检索的技能来指导其展开，而技能库则利用策略生成的有价值展开进行演化。这创造了一个闭环，其中改进的策略产生更好的技能，而不断发展的技能库又为策略展开提供更强的先验。跨越八个基准的实验表明，SPyCE始终优于基于强化学习和基于记忆的基线。进一步的分析表明，层次化技能设计和共同进化机制对我们的设计至关重要。这些结果表明，联合技能-策略优化是一种构建高效多模态智能体的有前景的范式。
cs.CL / 31 / 2607.13901
High-Order Question Generation in a Multilingual Educational Context
多语言教育背景下的高阶问题生成
Abstract
Critical thinking is a fundamental skill that helps learners move beyond simple memorization. One way to develop this skill is through high-order questioning. However, crafting such questions remains a challenge for educators, and classroom practices tend to rely on low-order questions. Large Language Models have demonstrated strong capabilities in generating high-order questions, especially when guided by prompts based on Bloom's Taxonomy. Yet, existing research has largely centered on this framework and focused only on English. This study addresses these gaps by introducing prompts grounded in two alternative frameworks: Claim-Evidence-Reasoning and Divergent Questioning within a multilingual context using Basque, Spanish, and English. Results indicate that while both an open-source and a proprietary model rather effectively generate questions in all three languages, only about half of the answerable questions are recognized by teachers as high-order. A positive finding is that the alternative frameworks produce structurally and conceptually varied questions, suggesting they could complement each other and provide viable alternatives to Bloom's Taxonomy.
Chinese Translation
批判性思维是一项基本技能，帮助学习者超越简单的记忆。发展这一技能的一种方法是通过高阶提问。然而，设计此类问题对教育工作者来说仍然是一个挑战，课堂实践往往依赖于低阶问题。大型语言模型在生成高阶问题方面表现出强大的能力，尤其是在基于布鲁姆分类法的提示指导下。然而，现有研究主要集中在这一框架上，并仅关注英语。本研究通过引入基于两种替代框架的提示来填补这些空白：主张-证据-推理（Claim-Evidence-Reasoning）和发散性提问（Divergent Questioning），并在使用巴斯克语、西班牙语和英语的多语言背景下进行。结果表明，尽管开源模型和专有模型在所有三种语言中都能相对有效地生成问题，但只有大约一半的可回答问题被教师认定为高阶问题。一个积极的发现是，这些替代框架产生了结构和概念上各异的问题，表明它们可以相辅相成，并为布鲁姆分类法提供可行的替代方案。
cs.CL / 32 / 2607.13920
DeepStress: Stress-Testing Deep Search Agents
DeepStress：深度搜索代理的压力测试
Abstract
While search agents demonstrate impressive capabilities in multi-step question answering, their robustness to poor-quality evidence remains under-explored. This phenomenon occurs rarely in realistic benchmarks but can lead to dramatic failure in real life applications. Therefore in this study we propose DeepStress, a stress testing framework that controls the frequency of challenging evidence by replacing the retrieval module of search agents with a controlled synthetic environment. We use this framework to control three dimensions that can affect document reliability: trustworthiness, relevance, and factuality. Testing several search agents on HotpotQA and BrowseCompPlus, we demonstrate that agents exhibit substantial differences in their ability to handle unreliable information and propose new metrics that better document systems outcomes as well as the interactions between conflicting parametric and retrieved knowledge.
Chinese Translation
尽管搜索代理在多步骤问答中表现出令人印象深刻的能力，但它们对低质量证据的鲁棒性仍然未得到充分探索。这种现象在现实基准测试中很少发生，但在实际应用中可能导致严重失败。因此，在本研究中，我们提出了DeepStress，一个压力测试框架，通过用受控的合成环境替换搜索代理的检索模块来控制具有挑战性的证据的频率。我们使用该框架控制三个可能影响文档可靠性的维度：可信度、相关性和事实性。在HotpotQA和BrowseCompPlus上测试多个搜索代理，我们证明了代理在处理不可靠信息的能力上存在显著差异，并提出了新的指标，以更好地记录系统结果以及冲突的参数知识和检索知识之间的相互作用。
cs.CL / 33 / 2607.13967
DeltaMerge-LowRes: Composing Language and Task Deltas for Low-Resource Adaptation
DeltaMerge-LowRes：为低资源适应组合语言和任务增量
Abstract
Adapting a multilingual encoder to a new language \emph{and} a new task with only a few hundred gold examples is a common low-resource NLP setting, yet the two axes are usually fused via an expensive language--task fine-tuning run. We ask whether they can instead be trained separately and recombined in weight space. \DeltaMergeLowRes{} learns a language delta $\Delta_L$ from unlabeled monolingual text and a task delta $\Delta_T$ from labeled English data, then composes them at inference under one of four rules: additive, activation-guided, sparsity-aware, and a novel \emph{cross-axis TIES}. The new rule adapts the TIES-Merging steps of trimming, sign election, and merging to the language and task axes rather than to two task axes. Holding $(\Delta_L,\Delta_T)$ fixed across rules on four task families and four African languages ($158$ evaluated cells, $10{,}000$-sample paired bootstrap per cell), we find: (i) cross-axis TIES wins summarisation on $3/4$ languages by $+4$ to $+7$ chrF (chrF $18.59$ vs.\ $13.80$ task-only); (ii) it improves QA F1 by $+2.32$ and EM by $+2.91$; and (iii) sparsity-aware merging cuts classification ECE by $36\%$ at parity macro-F1. The composition rule materially changes what the merged model preserves, suppresses, and calibrates. We release all JSON traces and a claim ledger.
Chinese Translation
将多语言编码器适应到一种新语言和一种新任务，仅使用几百个标注样本，是一种常见的低资源自然语言处理场景。然而，这两个维度通常通过昂贵的语言-任务微调过程融合在一起。我们探讨是否可以将它们分别训练并在权重空间中重新组合。DeltaMerge-LowRes 从未标注的单语文本中学习语言增量 $
abla_L$，并从标注的英语数据中学习任务增量 $
abla_T$，然后在推理时根据四种规则之一进行组合：加法、激活引导、稀疏感知，以及一种新颖的交叉轴 TIES。新规则将 TIES 合并步骤中的修剪、符号选择和合并适应到语言和任务轴，而不是两个任务轴。在四个任务类别和四种非洲语言上固定 $(
abla_L,
abla_T)$，我们发现：（i）交叉轴 TIES 在 $3/4$ 的语言上以 $+4$ 到 $+7$ 的 chrF（chrF $18.59$ 对比 $13.80$ 仅任务）赢得摘要任务；（ii）它将问答 F1 提高了 $+2.32$，EM 提高了 $+2.91$；（iii）稀疏感知合并在宏观 F1 相等的情况下将分类 ECE 降低了 $36 ext{ extperthousand}$。组合规则实质性地改变了合并模型所保留、抑制和校准的内容。我们发布了所有 JSON 跟踪和索赔账本。
cs.CL / 34 / 2607.13977
Constraint-Aware Counterfactual Editing for Aspect-Based Sentiment Analysis
面向约束的反事实编辑用于基于方面的情感分析
Abstract
Aspect-Based Sentiment Analysis (ABSA) requires models to identify sentiment toward specific aspects rather than relying on the global polarity of a sentence. This makes counterfactual evaluation especially challenging: a valid counterfactual should flip the sentiment of one target aspect while preserving the sentiment of all non-target aspects, semantic meaning, fluency, and factual consistency. Existing counterfactual generation methods often focus on sentence-level label flipping and may produce edits that are fluent but aspect-invalid, semantically drifting, or contradictory. To address this limitation, we propose CAVE-ABSA, a Constraint-Aware Validated Editing framework for generating and validating aspect-level counterfactuals. CAVE-ABSA localizes the opinion span associated with the target aspect, performs controlled counterfactual rewriting, refines candidates through a repair module, and filters them using aspect-level verification, semantic similarity, AMR-guided structural preservation, edit minimality, fluency, and contradiction detection. The framework is designed to construct validated counterfactual ABSA datasets for robustness evaluation and data augmentation. By explicitly separating generation from validation, CAVE-ABSA provides a principled approach for producing meaningful aspect-local counterfactuals and for testing whether ABSA models truly rely on aspect-grounded sentiment reasoning.
Chinese Translation
基于方面的情感分析（ABSA）要求模型识别特定方面的情感，而不是依赖句子的整体极性。这使得反事实评估尤其具有挑战性：有效的反事实应当翻转一个目标方面的情感，同时保持所有非目标方面的情感、语义意义、流畅性和事实一致性。现有的反事实生成方法通常侧重于句子级标签翻转，可能会产生流畅但方面无效、语义漂移或矛盾的编辑。为了解决这一局限性，我们提出了CAVE-ABSA，一个面向约束的验证编辑框架，用于生成和验证方面级反事实。CAVE-ABSA定位与目标方面相关的意见范围，执行受控的反事实重写，通过修复模块细化候选项，并使用方面级验证、语义相似性、AMR引导的结构保持、编辑最小化、流畅性和矛盾检测进行过滤。该框架旨在构建经过验证的反事实ABSA数据集，以进行鲁棒性评估和数据增强。通过明确将生成与验证分开，CAVE-ABSA提供了一种原则性的方法，用于生成有意义的方面局部反事实，并测试ABSA模型是否真正依赖于基于方面的情感推理。
cs.CL / 35 / 2607.14040
Can an Old Dog Be Taught New Tricks? Taking LLMs Beyond Sentence Level Translation
老狗能学新把戏吗？将大型语言模型推向超越句子级翻译的领域
Abstract
Automatic translation systems, from CAT tools to MT, overwhelmingly treat translation as a sentence-by-sentence act. This paper asks whether LLMs can be moved beyond that paradigm through whole-document, corpus-informed translation. We present PAT (Pragmatic Auto-Translator), a RAG-based system that pairs user-configured specifications with context from a comparable corpus of authentic longform texts in U.S. English and Latin American Spanish, passing retrieved paragraph-, section-, and document-level examples to an LLM for whole-document generation. The goal is draft translation for professional verification: target texts reformulated to fit their Spanish-language context, where discourse organization, rhetorical style, and pragmatic norms differ meaningfully from English. We evaluated six automatic translations of essays on generative AI across three projects using a customized MQM typology, assessed by two trained evaluators working from U.S. English into LATAM and Mexican Spanish. Results show that a limited prompt produced no meaningful reformulation, and specifications and corpus-informed translations at times showed substantial reformulation, though not always to effect. We find that LLMs can be moved toward reformulation and away from the sentence-by-sentence paradigm, though more work is needed to improve the effectiveness of those reformulations. In this paper, we discuss considerations related to automatic translation system design, corpus construction, and translation quality evaluation methodology and results.
Chinese Translation
自动翻译系统，从计算机辅助翻译工具到机器翻译，主要将翻译视为逐句进行的行为。本文探讨是否可以通过整篇文档和基于语料库的翻译，将大型语言模型（LLMs）推向这一范式之外。我们提出了PAT（Pragmatic Auto-Translator），一个基于检索增强生成（RAG）的系统，它将用户配置的规格与来自美国英语和拉丁美洲西班牙语的真实长文本的可比语料库中的上下文相结合，将检索到的段落、章节和文档级示例传递给LLM以进行整篇文档生成。目标是为专业验证提供草稿翻译：将目标文本重新表述以适应其西班牙语语境，其中话语组织、修辞风格和语用规范与英语有显著不同。我们使用定制的MQM类型学评估了三项项目中关于生成性人工智能的六个自动翻译的论文，由两名受过培训的评估人员从美国英语翻译成拉美和墨西哥西班牙语。结果表明，有限的提示未能产生有意义的重新表述，而规格和基于语料库的翻译在某些情况下显示出实质性的重新表述，尽管并不总是有效。我们发现，LLMs可以朝着重新表述的方向发展，摆脱逐句翻译的范式，尽管仍需更多工作来提高这些重新表述的有效性。本文讨论了与自动翻译系统设计、语料库构建以及翻译质量评估方法和结果相关的考虑因素。
cs.CL / 36 / 2607.14051
Hindcast: Replaying Prediction Markets to Evaluate LLM Forecasters
回溯预测：重放预测市场以评估大型语言模型预报员
Abstract
Forecasters are evaluated by backtesting, which replays resolved questions and grades the probability the system would have assigned before the outcome was known. For LLMs, two channels leak the answer into this test. A model that retrieves can surface reports written after the event, turning forecasting into a lookup, and each new model is trained on data closer to the event, so a question that lay in the future for last year's models sits inside this year's training data. Either way, the test grades recall while claiming to grade foresight. We introduce Hindcast, which closes both leaks by grading a model as if it stood at a chosen past date $t_0$, before the outcome existed in either channel. Hindcast replays resolved Polymarket prediction markets against a frozen snapshot of public Reddit, lets the model read only posts written before $t_0$, and scores each forecast against both what happened and the market's own price at $t_0$, itself a human forecast made from the same past information. Because the cutoff is set per market and the snapshot never changes, the evaluation re-runs on new markets as models improve, without going stale. Once the leak is closed, retrieval still helps most models, but only where Reddit discussed the event beforehand. Where the archive carried only speculation, retrieval hurts.
Chinese Translation
预报员的评估通过回测进行，该过程重放已解决的问题，并评估系统在结果已知之前所分配的概率。对于大型语言模型（LLMs），有两个渠道将答案泄露到这一测试中。一个能够检索的模型可能会调取事件发生后撰写的报告，将预测转变为查找，而每个新模型的训练数据都更接近事件，因此去年的模型所面临的未来问题在今年的训练数据中已经存在。无论如何，这一测试评估的是回忆能力，而声称评估的是前瞻性。我们引入了回溯预测（Hindcast），通过将模型评估为在某个选定的过去日期 $t_0$ 的状态，从而关闭了这两个泄露渠道，在此之前结果在任何渠道中都不存在。回溯预测重放已解决的 Polymarket 预测市场，并与公共 Reddit 的静态快照进行对比，让模型仅能读取 $t_0$ 之前撰写的帖子，并根据实际发生的情况和 $t_0$ 时市场自身的价格对每个预测进行评分，后者本身也是基于相同过去信息的人工预测。由于截止点是针对每个市场设定的，且快照从未改变，因此随着模型的改进，评估可以在新市场上重新进行，而不会过时。一旦泄露被关闭，检索仍然对大多数模型有帮助，但仅在 Reddit 事先讨论了该事件的情况下。如果档案中仅包含猜测，检索则会产生负面影响。