机器人 / 具身智能 - arXivDaily 专题

2509.11240 2025-09-16 cs.RO cs.SY eess.SY 57%

CORB-Planner: Corridor as Observations for RL Planning in High-Speed Flight

Yechen Zhang, Bin Gao, Gang Wang, Jian Sun, Zhuo Li

机构 * State Key Laboratory of Autonomous Intelligent Unmanned Systems, School of Automation, Beijing Institute of Technology（自主智能无人系统国家重点实验室，自动化学院，北京理工大学）

专题命中模仿学习与强化学习：robotic(abstract)；分类 cs.RO

Comments 11 pages, 8 figures. Submitted to IEEE/ASME T-MECH. Code available at https://github.com/ChenzycBIT/CORB-planner

2509.09772 2025-09-15 cs.LG stat.AP 57%

Hybrid Adaptive Conformal Offline Reinforcement Learning for Fair Population Health Management

Sanjay Basu, Sadiq Y. Patel, Parth Sheth, Bhairavi Muralidharan, Namrata Elamaran, Aakriti Kinra, Rajaie Batniji

机构 * Waymark, San Francisco, CA, USA（Waymark）； San Francisco General Hospital, University of California San Francisco, San Francisco, CA, USA（旧金山通用医院，加州大学旧金山分校）； University of California San Francisco, San Francisco, CA, USA（加州大学旧金山分校）； University of Pennsylvania, Philadelphia, PA, USA（宾夕法尼亚大学）

专题命中模仿学习与强化学习：navigation(abstract)；分类 cs.LG

Comments 10 pages, 5 figures, 4 tables

2509.08500 2025-09-11 cs.AI 57%

TCPO: Thought-Centric Preference Optimization for Effective Embodied Decision-making

Kechen Jiao, Zhirui Fang, Jiahao Liu, Bei Li, Qifan Wang, Xinyu Liu, Junhao Ruan, Zhongjian Qiao, Yifan Zhu, Yaxin Xu, Jingang Wang, Xiu Li

机构 * Tsinghua University（清华大学）； Meituan（美团）； Northeastern University（东北大学）； Beijing University of Posts and Telecommunications（北京邮电大学）； Meta AI ； Wuhan University（武汉大学）

专题命中模仿学习与强化学习：embodied agent(abstract)；分类 cs.AI

2410.00757 2025-09-09 cs.RO 57%

Collaborative motion planning for multi-manipulator systems through Reinforcement Learning and Dynamic Movement Primitives

Siddharth Singh, Tian Xu, Qing Chang

机构 * Department of Mechanical & Aerospace Engineering, University of Virginia（机械与航空航天工程系，弗吉尼亚大学）

专题命中模仿学习与强化学习：robotic(abstract)；分类 cs.RO

Comments 7 pages, 7 figures, conference submission

2509.04737 2025-09-08 cs.RO 57%

Imitation Learning Based on Disentangled Representation Learning of Behavioral Characteristics

Ryoga Oishi, Sho Sakaino, Toshiaki Tsuji

机构 * Saitama University（埼玉大学）； University of Tsukuba（筑波大学）

专题命中模仿学习与强化学习：robot learning(abstract)；分类 cs.RO

Comments 16 pages, 5 figures, Accepted at CoRL2025

2509.04259 2025-09-05 cs.LG 57%

RL's Razor: Why Online Reinforcement Learning Forgets Less

Idan Shenfeld, Jyothish Pari, Pulkit Agrawal

机构 * Improbable AI Lab, MIT（Improbable AI Lab 和马萨诸塞理工学院）

专题命中模仿学习与强化学习：robotic(abstract)；分类 cs.LG

2509.02164 2025-09-03 cs.CV 57%

Omnidirectional Spatial Modeling from Correlated Panoramas

Xinshen Zhang, Tongxi Fu, Xu Zheng

机构 * The Hong Kong Polytechnic University（香港理工大学）； Zhejiang Sci-Tech University（浙江科技学院）

专题命中模仿学习与强化学习：embodied AI(abstract)；分类 cs.CV

2508.21501 2025-09-01 cs.RO 57%

Few-Shot Neuro-Symbolic Imitation Learning for Long-Horizon Planning and Acting

Pierrick Lorang, Hong Lu, Johannes Huemer, Patrik Zips, Matthias Scheutz

机构 * Tufts University, United States（美国塔夫茨大学）； Austrian Institute of Technology, Austria（奥地利技术研究所）； Human-Robot Interaction Lab（人机交互实验室）

专题命中模仿学习与强化学习：robotic(abstract)；分类 cs.RO

Comments Accepted at CoRL 2025; to appear in PMLR

2508.16942 2025-08-26 cs.CV 57%

HieroAction: Hierarchically Guided VLM for Fine-Grained Action Analysis

Junhao Wu, Xiuer Gu, Zhiying Li, Yeying Jin, Yunfeng Diao, Zhiyu Li, Zhenbo Song, Xiaomei Zhang, Zhaoxin Fan

专题命中模仿学习与强化学习：robotics(abstract)；分类 cs.CV

2508.15013 2025-08-22 cs.AI q-bio.NC 57%

Goals and the Structure of Experience

Nadav Amir, Stas Tiomkin, Angela Langdon

专题命中模仿学习与强化学习：world model(abstract)；分类 cs.AI

2406.13434 2025-08-15 cs.RO 57%

Tactile Aware Dynamic Obstacle Avoidance in Crowded Environment with Deep Reinforcement Learning

Yung Chuen Ng, Qi Wen Shervina Lim, Chun Ye Tan, Zhen Hao Gan, Meng Yee Michael Chuah

机构 * Institute for Infocomm Research (I 2 R), A*STAR（信息通信研究所（I2R），A*STAR）； Nanyang Technological University (NTU)（南洋理工大学）

专题命中模仿学习与强化学习：navigation(abstract)；分类 cs.RO

2409.08511 2025-08-14 cs.RO 57%

Vision-driven UAV River Following: Benchmarking with Safe Reinforcement Learning

Zihan Wang, Nina Mahmoudian

机构 * School of Mechanical Engineering, Purdue University（机械工程学院，普渡大学）

专题命中模仿学习与强化学习：navigation(abstract)；分类 cs.RO

Comments Accepted by conference IFAC CAMS 2024

Journal ref IFAC-PapersOnLine, 58(20): 421-427, 2024

2401.09332 2025-08-14 cs.RO 57%

Synergistic Reinforcement and Imitation Learning for Vision-driven Autonomous Flight of UAV Along River

Zihan Wang, Jianwen Li, Nina Mahmoudian

机构 * School of Mechanical Engineering, Purdue University（机械工程学院，普渡大学）

专题命中模仿学习与强化学习：navigation(abstract)；分类 cs.RO

Comments Submitted to IROS2024

Journal ref 979-8-3503-7770-5/24/$31.00 \c{opyright}2024 IEEE

2410.15250 2025-08-13 cs.LG 57%

Multi-modal Policies with Physics-informed Representations in Complex Fluid Environments

Haodong Feng, Peiyan Hu, Yue Wang, Dixia Fan

专题命中模仿学习与强化学习：robotics(abstract)；分类 cs.LG

2503.08029 2025-08-12 cs.RO cs.SY eess.SY 57%

Elastic Motion Policy: An Adaptive Dynamical System for Robust and Efficient One-Shot Imitation Learning

Tianyu Li, Sunan Sun, Shubhodeep Shiv Aditya, Nadia Figueroa

机构 * School of Engineering and Applied Science, University of Pennsylvania（工程与应用科学学院，宾夕法尼亚大学）

专题命中模仿学习与强化学习：robotics(abstract)；分类 cs.RO

2508.06804 2025-08-12 cs.RO 57%

D3P: Dynamic Denoising Diffusion Policy via Reinforcement Learning

Shu-Ang Yu, Feng Gao, Yi Wu, Chao Yu, Yu Wang

专题命中模仿学习与强化学习：robotic(abstract)；分类 cs.RO

2508.06330 2025-08-11 cs.RO 57%

L2Calib: $SE(3)$-Manifold Reinforcement Learning for Robust Extrinsic Calibration with Degenerate Motion Resilience

Baorun Li, Chengrui Zhu, Siyi Du, Bingran Chen, Jie Ren, Wenfei Wang, Yong Liu, Jiajun Lv

机构 * Institute of Cyber-Systems and Control, Zhejiang University, China（浙江大学控制科学与工程学院）； State Key Laboratory of Industrial Control Technology, Zhejiang University, China（浙江大学工业控制技术状态重点实验室）； Zhejiang Guozi Robotics Technology Co., Ltd.（浙江国智机器人科技有限公司）

专题命中模仿学习与强化学习：robotic(abstract)；分类 cs.RO

Comments IROS2025

2508.05619 2025-08-08 cs.AI nlin.AO physics.bio-ph physics.comp-ph physics.hist-ph 57%

The Missing Reward: Active Inference in the Era of Experience

Bo Wen

机构 * IBM T.J. Watson Research Center（IBM TJ沃森研究中心）

专题命中模仿学习与强化学习：world model(abstract)；分类 cs.AI

2508.03910 2025-08-07 q-fin.CP cs.LG 57%

Comparing Normalization Methods for Portfolio Optimization with Reinforcement Learning

Caio de Souza Barbosa Costa, Anna Helena Reali Costa

机构 * Escola Politécnica Universidade de São Paulo (USP)（圣保罗大学理工学院）

专题命中模仿学习与强化学习：robotics(abstract)；分类 cs.LG

2508.03012 2025-08-07 cs.SE cs.AI 57%

Tool-integrated Reinforcement Learning for Repo Deep Search

Zexiong Ma, Chao Peng, Qunhong Zeng, Pengfei Gao, Yanzhen Zou, Bing Xie

机构 * Peking University（北京大学）； ByteDance（字节跳动）； Beijing Institute of Technology（北京理工大学）

专题命中模仿学习与强化学习：navigation(abstract)；分类 cs.AI

2310.05858 2025-08-07 cs.LG cs.SY eess.SY 57%

Distributional Soft Actor-Critic with Three Refinements

Jingliang Duan, Wenxuan Wang, Liming Xiao, Jiaxin Gao, Shengbo Eben Li, Chang Liu, Ya-Qin Zhang, Bo Cheng, Keqiang Li

机构 * School of Mechanical Engineering, University of Science and Technology Beijing（北京科技大学机械工程学院）； School of Vehicle and Mobility, Tsinghua University（清华大学车辆与移动性学院）； College of Engineering, Peking University（北京大学工程学院）； Institute for AI Industry Research, Tsinghua University（清华大学人工智能产业研究院）

专题命中模仿学习与强化学习：robotic(abstract)；分类 cs.LG

Comments Title updated in this version. The previous version was titled "DSAC-T: Distributional Soft Actor-Critic With Three Refinements". Added a footnote with the BibTeX entry for the published journal version. No other major changes

Journal ref IEEE Trans. PAMI,47(5): 3935-3946,2025

2508.03129 2025-08-06 cs.RO 57%

Safety-Aware Imitation Learning via MPC-Guided Disturbance Injection

Le Qiu, Yusuf Umut Ciftci, Somil Bansal

机构 * Department of Electrical Engineering, Tsinghua University（电子工程系，清华大学）； Department of Electrical and Computer Engineering, University of Southern California（电气与计算机工程系，南加州大学）； Department of Aeronautics and Astronautics, Stanford University（航空与航天工程系，斯坦福大学）

专题命中模仿学习与强化学习：navigation(abstract)；分类 cs.RO

2406.14476 2025-08-06 cs.AI 57%

Learning telic-controllable state representations

Nadav Amir, Stas Tiomkin

机构 * Princeton Neuroscience Institute, Princeton University（普林斯顿神经科学研究所，普林斯顿大学）； Department of Computer Science, Texas Tech University（计算机科学系，德克萨斯技术大学）

专题命中模仿学习与强化学习：navigation(abstract)；分类 cs.AI

Comments Published in Proceedings of the 47th Annual Meeting of the Cognitive Science Society

Journal ref Proceedings of the Annual Meeting of the Cognitive Science Society, 47 (2025)

2505.04974 2025-08-04 cs.CV 57%

ReAlign: Bilingual Text-to-Motion Generation via Step-Aware Reward-Guided Alignment

Wanjiang Weng, Xiaofeng Tan, Hongsong Wang, Pan Zhou

专题命中模仿学习与强化学习：robotics(abstract)；分类 cs.CV

Comments We believe that there are some areas in the manuscript that require further improvement, and out of our commitment to refining this work, we have decided to withdraw our manuscript after careful deliberation and discussion

2507.23562 2025-08-01 cs.LG cs.AR 57%

Hardware-Aware Fine-Tuning of Spiking Q-Networks on the SpiNNaker2 Neuromorphic Platform

Sirine Arfa, Bernhard Vogginger, Christian Mayr

机构 * Centre for Tactile Internet with Human-in-the-Loop (CeTI)（触觉互联网与人环路中心）

专题命中模仿学习与强化学习：robotic(abstract)；分类 cs.LG

Comments 8 pages, 5 figures, 3 tables

Journal ref ACM ICONS 2025 - International Conference on Neuromorphic Systems

2504.06684 2025-08-01 cs.RO cs.MA 57%

SDHN: Skewness-Driven Hypergraph Networks for Enhanced Localized Multi-Robot Coordination

Delin Zhao, Yanbo Shan, Chang Liu, Shenghang Lin, Yingxin Shou, Bin Xu

机构 * School of Automation, Northwestern Polytechnical University, Xi'an, 710072, China（西北工业大学自动化学院）； School of Automation, Southeast University, Nanjing, 210096, China（东南大学自动化学院）

专题命中模仿学习与强化学习：robotic(abstract)；分类 cs.RO

2503.09722 2025-07-29 cs.LG cs.SY eess.SY stat.ML 57%

The Pitfalls of Imitation Learning when Actions are Continuous

Max Simchowitz, Daniel Pfrommer, Ali Jadbabaie

机构 * CMU（卡内基梅隆大学）； MIT（麻省理工学院）

专题命中模仿学习与强化学习：robot learning(abstract)；分类 cs.LG

Comments 98 pages, 2 figures, updated proof sketch

2507.18113 2025-07-25 cs.LG 57%

Policy Disruption in Reinforcement Learning:Adversarial Attack with Large Language Models and Critical State Identification

Junyong Jiang, Buwei Tian, Chenxing Xu, Songze Li, Lu Dong

机构 * School of Cyber Science and Engineering（网络科学与工程学院）

专题命中模仿学习与强化学习：robotics(abstract)；分类 cs.LG

2507.16941 2025-07-24 cs.RO 57%

Multi-agent Reinforcement Learning for Robotized Coral Reef Sample Collection

Daniel Correa, Tero Kaarlela, Jose Fuentes, Paulo Padrao, Alain Duran, Leonardo Bobadilla

机构 * Knight Foundation School of Computing and Information Sciences, Florida International University（凯尼恩基金会计算与信息科学学院，佛罗里达国际大学）； College of Arts, Sciences & Education, Biological Sciences, Florida International University（艺术、科学与教育学院，生物科学，佛罗里达国际大学）

专题命中模仿学习与强化学习：robotic(abstract)；分类 cs.RO

2507.15073 2025-07-22 cs.LG 57%

Reinforcement Learning for Flow-Matching Policies

Samuel Pfrommer, Yixiao Huang, Somayeh Sojoudi

机构 * Department of Electrical Engineering and Computer Sciences University of California, Berkeley（电气工程与计算机科学系加州大学伯克利分校）

专题命中模仿学习与强化学习：robotics(abstract)；分类 cs.LG

1. 模仿学习与强化学习 4134 篇

CORB-Planner: Corridor as Observations for RL Planning in High-Speed Flight

Hybrid Adaptive Conformal Offline Reinforcement Learning for Fair Population Health Management

TCPO: Thought-Centric Preference Optimization for Effective Embodied Decision-making

Collaborative motion planning for multi-manipulator systems through Reinforcement Learning and Dynamic Movement Primitives

Imitation Learning Based on Disentangled Representation Learning of Behavioral Characteristics

RL's Razor: Why Online Reinforcement Learning Forgets Less

Omnidirectional Spatial Modeling from Correlated Panoramas

Few-Shot Neuro-Symbolic Imitation Learning for Long-Horizon Planning and Acting

HieroAction: Hierarchically Guided VLM for Fine-Grained Action Analysis

Goals and the Structure of Experience

Tactile Aware Dynamic Obstacle Avoidance in Crowded Environment with Deep Reinforcement Learning

Vision-driven UAV River Following: Benchmarking with Safe Reinforcement Learning

Synergistic Reinforcement and Imitation Learning for Vision-driven Autonomous Flight of UAV Along River

Multi-modal Policies with Physics-informed Representations in Complex Fluid Environments

Elastic Motion Policy: An Adaptive Dynamical System for Robust and Efficient One-Shot Imitation Learning

D3P: Dynamic Denoising Diffusion Policy via Reinforcement Learning

L2Calib: $SE(3)$-Manifold Reinforcement Learning for Robust Extrinsic Calibration with Degenerate Motion Resilience

The Missing Reward: Active Inference in the Era of Experience

Comparing Normalization Methods for Portfolio Optimization with Reinforcement Learning

Tool-integrated Reinforcement Learning for Repo Deep Search

Distributional Soft Actor-Critic with Three Refinements

Safety-Aware Imitation Learning via MPC-Guided Disturbance Injection

Learning telic-controllable state representations

ReAlign: Bilingual Text-to-Motion Generation via Step-Aware Reward-Guided Alignment

Hardware-Aware Fine-Tuning of Spiking Q-Networks on the SpiNNaker2 Neuromorphic Platform

SDHN: Skewness-Driven Hypergraph Networks for Enhanced Localized Multi-Robot Coordination

The Pitfalls of Imitation Learning when Actions are Continuous

Policy Disruption in Reinforcement Learning:Adversarial Attack with Large Language Models and Critical State Identification

Multi-agent Reinforcement Learning for Robotized Coral Reef Sample Collection

Reinforcement Learning for Flow-Matching Policies