机器人 / 具身智能 - arXivDaily 专题

2510.08218 2025-10-10 cs.LG cs.AI 62%

Expressive Value Learning for Scalable Offline Reinforcement Learning

Nicolas Espinosa-Dice, Kiante Brantley, Wen Sun

机构 * Department of Computer Science（计算机科学系）； Cornell University（康奈尔大学）； Kempner Institute（凯普纳研究所）； Harvard University（哈佛大学）

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

Comments 24 pages, 5 figures

2510.03578 2025-10-07 cs.LG cs.AI stat.ML 62%

Latent Mixture of Symmetries for Sample-Efficient Dynamic Learning

Haoran Li, Chenhan Xiao, Muhao Guo, Yang Weng

机构 * School of Electrical, Computer and Energy Engineering（电气、计算机与能源工程学院）； Arizona State University（亚利桑那州立大学）

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

Comments 30 pages, 6 figures

2510.03302 2025-10-07 cs.LG cs.CV 62%

Revoking Amnesia: RL-based Trajectory Optimization to Resurrect Erased Concepts in Diffusion Models

Daiheng Gao, Nanxiang Jiang, Andi Zhang, Shilin Lu, Yufei Tang, Wenbo Zhou, Weiming Zhang, Zhaoxin Fan

机构 * USTC（中国科学技术大学）； Beihang University（北京航空航天大学）； University of Manchester（曼彻斯特大学）； NTU（南洋理工大学）； FYUST（福建师范大学）

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

Comments 21 pages, 10 figures

2503.02075 2025-10-06 cs.RO cs.LG cs.SY eess.SY 62%

Active Alignments of Lens Systems with Reinforcement Learning

Matthias Burkhardt, Tobias Schmähling, Pascal Stegmann, Michael Layh, Tobias Windisch

机构 * Institute for Machine Vision, University of Applied Sciences Kempten（机器视觉研究所，应用科技大学凯普腾）

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

Comments This work has been submitted to the IEEE for possible publication

2509.20766 2025-09-30 cs.RO cs.LG 62%

Leveraging Temporally Extended Behavior Sharing for Multi-task Reinforcement Learning

Gawon Lee, Daesol Cho, H. Jin Kim

机构 * Department of Aerospace Engineering, Seoul National University（首尔国立大学航空航天工程系）； Artificial Intelligence Institute, Seoul National University（首尔国立大学人工智能研究所）

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

Comments Accepted for publication in the proceedings of the 2025 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

2509.06775 2025-09-30 eess.SY cs.AI cs.IT cs.LG cs.NI cs.SY math.IT 62%

Agentic DDQN-Based Scheduling for Licensed and Unlicensed Band Allocation in Sidelink Networks

Po-Heng Chou, Pin-Qi Fu, Walid Saad, Li-Chun Wang

机构 * Department of Electronics and Electrical Engineering（电子与电气工程系）； Institute of Communications Engineering（通讯工程研究所）； National Yang Ming Chiao Tung University（阳明交通大学）； Research Center for Information Technology Innovation（信息技术创新研究中心）； Academia Sinica（台湾“中央研究院”）； Bradley Department of Electrical and Computer Engineering（电气与计算机工程系）

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

Comments 6 pages, 3 figures, accepted by 2025 IEEE Globecom Workshops

2506.08440 2025-09-30 cs.RO cs.AI 62%

TGRPO :Fine-tuning Vision-Language-Action Model via Trajectory-wise Group Relative Policy Optimization

Zengjue Chen, Runliang Niu, He Kong, Qi Wang, Qianli Xing, Zipei Fan

机构 * School of Artificial Intelligence, Jilin University（人工智能学院，吉林大学）

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

2505.10482 2025-09-30 cs.LG cs.AI 62%

Fine-tuning Diffusion Policies with Backpropagation Through Diffusion Timesteps

Ningyuan Yang, Jiaxuan Gao, Feng Gao, Yi Wu, Chao Yu

机构 * Tsinghua University（清华大学）

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

Comments 9 pages for main text, 23 pages in total, submitted to Neurips, 13 figures

2503.12243 2025-09-30 cs.RO cs.AI 62%

RISE: Robust Imitation through Stochastic Encoding

Mumuksh Tayal, Manan Tayal, Ravi Prakash

机构 * Cyber Physical Systems, Indian Institute of Science (IISc), Bengaluru（印度科学研究院计算机物理系统，班加罗尔）

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

Comments 15 pages, 4 figures

2509.22130 2025-09-29 cs.MA cs.AI cs.LG 62%

Multi-Agent Path Finding via Offline RL and LLM Collaboration

Merve Atasever, Matthew Hong, Mihir Nitin Kulkarni, Qingpei Li, Jyotirmoy V. Deshmukh

机构 * Department of Computer Science, University of Southern California（计算机科学系，南加州大学）

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

2509.19301 2025-09-29 cs.RO cs.LG 62%

Residual Off-Policy RL for Finetuning Behavior Cloning Policies

Lars Ankile, Zhenyu Jiang, Rocky Duan, Guanya Shi, Pieter Abbeel, Anusha Nagabandi

机构 * Amazon FAR (Frontier AI & Robotics)（亚马逊前沿人工智能与机器人实验室）； Stanford University（斯坦福大学）； Carnegie Mellon University（卡内基梅隆大学）； UC Berkeley（加州大学伯克利分校）

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

Comments Project website: https://residual-offpolicy-rl.github.io

2505.15173 2025-09-24 cs.CV cs.AI 62%

AvatarShield: Visual Reinforcement Learning for Human-Centric Synthetic Video Detection

Zhipei Xu, Xuanyu Zhang, Qing Huang, Xing Zhou, Jian Zhang

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

2509.17274 2025-09-23 cs.RO cs.LG math.OC 62%

Learning and Optimization with 3D Orientations

Alexandros Ntagkas, Constantinos Tsakonas, Chairi Kiourt, Konstantinos Chatzilygeroudis

机构 * Computational Intelligence Laboratory (CILab), Department of Mathematics, University of Patras（计算智能实验室（CILab），数学系，帕特拉大学）； Laboratory of Automation and Robotics (LAR) in the Department of Electrical & Computer Engineering, University of Patras（自动化与机器人实验室（LAR），电气与计算机工程系，帕特拉大学）； Archimedes/Athena RC, Greece（阿基米德/雅典RC，希腊）； Athena - Research and Innovation Center in Information, Communication and Knowledge Technologies, Xanthi, Greece（雅典-信息、通信和知识技术研究与创新中心，辛提，希腊）

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

Comments 9 pages, 11 figures

2509.11367 2025-09-16 cs.LG cs.AI 62%

Detecting Model Drifts in Non-Stationary Environment Using Edit Operation Measures

Chang-Hwan Lee, Alexander Shim

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

Comments 28 pages, 3 figures, 17 tables

2509.08257 2025-09-12 cs.RO cs.AI 62%

Symmetry-Guided Multi-Agent Inverse Reinforcement Learning

Yongkai Tian, Yirong Qi, Xin Yu, Wenjun Wu, Jie Luo

机构 * State Key Laboratory of Complex & Critical Software Environment, Beihang University, Beijing, China（复杂与关键软件环境国家重点实验室，北京航空航天大学）

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

Comments 8pages, 6 figures. Accepted for publication in the Proceedings of the 2025 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2025) as oral presentation

2509.05735 2025-09-09 cs.LG cs.AI 62%

Offline vs. Online Learning in Model-based RL: Lessons for Data Collection Strategies

Jiaqi Chen, Ji Shi, Cansu Sancaktar, Jonas Frey, Georg Martius

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

Comments Accepted at Reinforcement Learning Conference (RLC 2025); Code available at: https://github.com/swsychen/Offline_vs_Online_in_MBRL

2509.04628 2025-09-08 cs.RO cs.AI 62%

Action Chunking with Transformers for Image-Based Spacecraft Guidance and Control

Alejandro Posadas-Nava, Andrea Scorsoglio, Luca Ghilardi, Roberto Furfaro, Richard Linares

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

Comments 12 pages, 6 figures, 2025 AAS/AIAA Astrodynamics Specialist Conference

2503.22524 2025-09-05 cs.RO cs.AI 62%

Robust Offline Imitation Learning Through State-level Trajectory Stitching

Shuze Wang, Yunpeng Mei, Hongjie Cao, Yetian Yuan, Gang Wang, Jian Sun, Jie Chen

机构 * National Key Lab of Autonomous Intelligent Unmanned Systems, Beijing Institute of Technology, Beijing 100081, China（自主智能无人系统国家重点实验室，北京理工大学，北京100081，中国）； Department of Control Science and Engineering, Harbin Institute of Technology, Harbin 150001, China（控制科学与工程学院，哈尔滨工业大学，哈尔滨150001，中国）

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

2509.00574 2025-09-03 cs.RO cs.LG 62%

Learning Dolly-In Filming From Demonstration Using a Ground-Based Robot

Philip Lorimer, Alan Hunter, Wenbin Li

机构 * Department of Computer Science, University of Bath, UK（计算机科学系，巴斯大学）

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

Comments Preprint; under double-anonymous review. 6 pages

2403.08955 2025-09-03 cs.LG cs.AI math.OC 62%

Towards Efficient Risk-Sensitive Policy Gradient: An Iteration Complexity Analysis

Rui Liu, Anish Gupta, Erfaun Noorani, Pratap Tokekar

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

2508.21101 2025-09-01 cs.LG cs.AI 62%

Beyond Prediction: Reinforcement Learning as the Defining Leap in Healthcare AI

Dilruk Perera, Gousia Habib, Qianyi Xu, Daniel J. Tan, Kai He, Erik Cambria, Mengling Feng

机构 * National University of Singapore（新加坡国立大学）； Nanyang Technological University（南洋理工大学）

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

Comments 40 pages in total (including appendix)

2504.15876 2025-08-28 cs.RO cs.AI 62%

Bidirectional Task-Motion Planning Based on Hierarchical Reinforcement Learning for Strategic Confrontation

Qizhen Wu, Lei Chen, Kexin Liu, Jinhu Lu

机构 * School of Automation Science and Electrical Engineering, Beihang University（自动化科学与电气工程学院，北京航空航天大学）； Advanced Research Institute of Multidisciplinary Sciences and State Key Laboratory of CNS/ATM, Beijing Institute of Technology（多学科科学先进研究院和 CNS/ATM 国家重点实验室，北京理工大学）

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

2302.12244 2025-08-28 cs.RO cs.LG 62%

To the Noise and Back: Diffusion for Shared Autonomy

Takuma Yoneda, Luzhe Sun, Ge Yang, Bradly Stadie, Matthew Walter

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

Comments https://diffusion-for-shared-autonomy.github.io/

2406.04920 2025-08-26 cs.RO cs.LG cs.SY eess.SY 62%

Sim-to-Real Transfer of Deep Reinforcement Learning Agents for Online Coverage Path Planning

Arvi Jonnarth, Ola Johansson, Jie Zhao, Michael Felsberg

机构 * Department of Electrical Engineering, Linköping University, Sweden（瑞典林雪平大学电气工程系）； Department of Information and Communication Engineering, Dalian University of Technology, China（中国大连理工大学信息与通信工程学院）； School of Engineering, University of KwaZulu-Natal, Durban, South Africa（南非夸祖鲁-纳塔尔大学工程学院）

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

Comments Published in IEEE Access

Journal ref IEEE Access, 2025, volume 13, pages 106883-106905

2410.07071 2025-08-14 cs.LG cs.AI 62%

Retrieval-Augmented Decision Transformer: External Memory for In-context RL

Thomas Schmied, Fabian Paischer, Vihang Patil, Markus Hofmarcher, Razvan Pascanu, Sepp Hochreiter

机构 * ELLIS Unit, LIT AI Lab, Institute for Machine Learning, JKU Linz, Austria（ELLIS单位、LIT人工智能实验室、机器学习研究所、JKU林茨、奥地利）； Extensity AI ； Google DeepMind（谷歌DeepMind）； Mila - Québec AI Institute（魁北克人工智能研究所）； NXAI

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

2404.07826 2025-08-12 cs.LG cs.AI 62%

On the Sample Efficiency of Abstractions and Potential-Based Reward Shaping in Reinforcement Learning

Giuseppe Canonaco, Leo Ardon, Alberto Pozanco, Daniel Borrajo

机构 * AI Research Dept. JPMorganChase Madrid, ES（摩根大通西班牙马德里人工智能研究部）； AI Research Dept. JPMorganChase London, UK（摩根大通英国伦敦人工智能研究部）

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

2508.04280 2025-08-07 cs.LG cs.AI 62%

Enhancing Vision-Language Model Training with Reinforcement Learning in Synthetic Worlds for Real-World Success

George Bredis, Stanislav Dereka, Viacheslav Sinii, Ruslan Rakhimov, Daniil Gavrilov

机构 * George Bredis（无）； Stanislav Dereka（无）； Viacheslav Sinii（无）； Ruslan Rakhimov（无）； Daniil Gavrilov（无）

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

2508.02219 2025-08-05 cs.RO cs.LG 62%

CO-RFT: Efficient Fine-Tuning of Vision-Language-Action Models through Chunked Offline Reinforcement Learning

Dongchi Huang, Zhirui Fang, Tianle Zhang, Yihang Li, Lin Zhao, Chunhe Xia

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

2503.05833 2025-08-05 cs.RO cs.LG 62%

Refined Policy Distillation: From VLA Generalists to RL Experts

Tobias Jülg, Wolfram Burgard, Florian Walter

机构 * Department of Computer Science & Artificial Intelligence, University of Technology Nuremberg（计算机科学与人工智能系，图恩大学）

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

Comments accepted for publication at IROS 2026

2503.15897 2025-08-01 cs.CV cs.LG 62%

Learning 3D Scene Analogies with Neural Contextual Scene Maps

Junho Kim, Gwangtak Bae, Eun Sun Lee, Young Min Kim

机构 * Dept. of Electrical and Computer Engineering, Seoul National University（电子与计算机工程系，首尔国立大学）； Interdisciplinary Program in Artificial Intelligence and INMC, Seoul National University（人工智能跨学科项目及INMC，首尔国立大学）

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

Comments Accepted to ICCV 2025

1. 模仿学习与强化学习 4126 篇

Expressive Value Learning for Scalable Offline Reinforcement Learning

Latent Mixture of Symmetries for Sample-Efficient Dynamic Learning

Revoking Amnesia: RL-based Trajectory Optimization to Resurrect Erased Concepts in Diffusion Models

Active Alignments of Lens Systems with Reinforcement Learning

Leveraging Temporally Extended Behavior Sharing for Multi-task Reinforcement Learning

Agentic DDQN-Based Scheduling for Licensed and Unlicensed Band Allocation in Sidelink Networks

TGRPO :Fine-tuning Vision-Language-Action Model via Trajectory-wise Group Relative Policy Optimization

Fine-tuning Diffusion Policies with Backpropagation Through Diffusion Timesteps

RISE: Robust Imitation through Stochastic Encoding

Multi-Agent Path Finding via Offline RL and LLM Collaboration

Residual Off-Policy RL for Finetuning Behavior Cloning Policies

AvatarShield: Visual Reinforcement Learning for Human-Centric Synthetic Video Detection

Learning and Optimization with 3D Orientations

Detecting Model Drifts in Non-Stationary Environment Using Edit Operation Measures

Symmetry-Guided Multi-Agent Inverse Reinforcement Learning

Offline vs. Online Learning in Model-based RL: Lessons for Data Collection Strategies

Action Chunking with Transformers for Image-Based Spacecraft Guidance and Control

Robust Offline Imitation Learning Through State-level Trajectory Stitching

Learning Dolly-In Filming From Demonstration Using a Ground-Based Robot

Towards Efficient Risk-Sensitive Policy Gradient: An Iteration Complexity Analysis

Beyond Prediction: Reinforcement Learning as the Defining Leap in Healthcare AI

Bidirectional Task-Motion Planning Based on Hierarchical Reinforcement Learning for Strategic Confrontation

To the Noise and Back: Diffusion for Shared Autonomy

Sim-to-Real Transfer of Deep Reinforcement Learning Agents for Online Coverage Path Planning

Retrieval-Augmented Decision Transformer: External Memory for In-context RL

On the Sample Efficiency of Abstractions and Potential-Based Reward Shaping in Reinforcement Learning

Enhancing Vision-Language Model Training with Reinforcement Learning in Synthetic Worlds for Real-World Success

CO-RFT: Efficient Fine-Tuning of Vision-Language-Action Models through Chunked Offline Reinforcement Learning

Refined Policy Distillation: From VLA Generalists to RL Experts

Learning 3D Scene Analogies with Neural Contextual Scene Maps