机器人 / 具身智能

X-Diffusion: Training Diffusion Policies on Cross-Embodiment Human Demonstrations

X-Diffusion：在跨具身人类示范上训练扩散策略

Maximus A. Pace, Prithwish Dan, Chuanruo Ning, Atiksh Bhardwaj, Audrey Du, Edward W. Duan, Wei-Chiu Ma, Kushal Kedia

机构 * Cornell University（康奈尔大学） 纠错

专题命中 机器人学习 ：robot learning(abstract)；manipulation(abstract)；分类 cs.RO、cs.AI、cs.CV

AI总结 本文提出X-Diffusion框架，通过在噪声人类动作上训练扩散策略，提升机器人任务成功率16%。

Comments ICRA 2026

Learning from Imperfect Demonstrations via Temporal Behavior Tree-Guided Trajectory Repair

通过时间行为树引导轨迹修复学习不完美的示范

Aniruddh G. Puranic, Sebastian Schirmer, John S. Baras, Calin Belta

专题命中 机器人学习 ：robot learning(abstract)；navigation(abstract)；分类 cs.RO、cs.AI、cs.LG

AI总结 本文提出利用时间行为树修复不完美示范轨迹，以提升机器人控制策略的学习效果，通过生成逻辑一致且可解释的数据集，指导强化学习 agent 向任务一致的 state space 区域发展。

Comments 12 pages, 4 figures. This work has been submitted to the IEEE for possible publication

Grounding Bodily Awareness in Visual Representations for Efficient Policy Learning

在视觉表示中 grounding 身体意识以实现高效的策略学习

Junlin Wang, Zhiyun Lin

机构 * SUSTech Shenzhen（深圳科技大学） 纠错

专题命中 机器人学习 ：manipulation(abstract)；robotic(abstract)；分类 cs.RO、cs.CV、cs.LG

AI总结 本文提出ICon方法，通过对比学习提升机器人操作中策略学习的效率和跨机器人迁移能力。

Comments A preprint version

End-to-end Optimization of Belief and Policy Learning in Shared Autonomy Paradigms

端到端优化信念与策略学习在共享自主性范式中的应用

MH Farhadi, Ali Rabiee, Sima Ghafoori, Anna Cetera, Andrew Fisher, Reza Abiri

机构 * University of Rhode Island（罗德岛大学） 纠错

专题命中 机器人学习 ：manipulation(abstract)；robotic(abstract)；分类 cs.RO、cs.AI、cs.LG

AI总结 BRACE框架通过整合信念信息与策略学习，提升了机器人在复杂目标模糊场景中的自主性和效率，实现了更高的成功率和路径效率。

FreqPolicy: Frequency Autoregressive Visuomotor Policy with Continuous Tokens

Yiming Zhong, Yumeng Liu, Chuyang Xiao, Zemin Yang, Youzhuo Wang, Yufei Zhu, Ye Shi, Yujing Sun, Xinge Zhu, Yuexin Ma

机构 * ShanghaiTech University（上海科技大学） ； The University of Hong Kong（香港大学） ； Digital Trust Centre, Nanyang Technological University（南洋理工大学数字信任中心） ； The Chinese University of Hong Kong（香港中文大学） 纠错

专题命中 机器人学习 ：manipulation(abstract)；robotic(abstract)；分类 cs.RO、cs.AI、cs.CV

Comments Comments: Published at Neural Information Processing Systems (NeurIPS) 2025. Project page and code: https://freq-policy.github.io/

In-Context Policy Adaptation via Cross-Domain Skill Diffusion

Minjong Yoo, Woo Kyung Kim, Honguk Woo

专题命中 机器人学习 ：manipulation(abstract)；robotic(abstract)；分类 cs.RO、cs.AI、cs.LG

Comments 9 pages

Journal ref Proceedings of the AAAI Conference on Artificial Intelligence, 2025

IMLE Policy: Fast and Sample Efficient Visuomotor Policy Learning via Implicit Maximum Likelihood Estimation

Krishan Rana, Robert Lee, David Pershouse, Niko Suenderhauf

专题命中 机器人学习 ：robotics(abstract)；manipulation(abstract)；分类 cs.RO、cs.AI、cs.LG

Comments Videos and code are available at https://imle-policy.github.io/

Learning from Demonstration with Implicit Nonlinear Dynamics Models

Peter David Fagan, Subramanian Ramamoorthy

专题命中 机器人学习 ：manipulation(abstract)；robotic(abstract)；分类 cs.RO、cs.AI、cs.LG

Comments 21 pages, 9 figures

Towards General Purpose Robots at Scale: Lifelong Learning and Learning to Use Memory

William Yue

专题命中 机器人学习 ：robotics(abstract)；robot learning(abstract)；分类 cs.RO、cs.AI、cs.LG

Comments arXiv admin note: substantial text overlap with arXiv:2401.02576, arXiv:2411.07954

Text-Aware Diffusion for Policy Learning

Calvin Luo, Mandy He, Zilai Zeng, Chen Sun

专题命中 机器人学习 ：manipulation(abstract)；robotic(abstract)；分类 cs.AI、cs.CV、cs.LG

Estimating Neural Network Robustness via Lipschitz Constant and Architecture Sensitivity

Abulikemu Abuduweili, Changliu Liu

专题命中 机器人学习 ：robot learning(abstract)；robotic(abstract)；分类 cs.RO、cs.AI、cs.LG

Comments SAFE-ROL at CoRL 2024

SkillDiffuser: Interpretable Hierarchical Planning via Skill Abstractions in Diffusion-Based Task Execution

Zhixuan Liang, Yao Mu, Hengbo Ma, Masayoshi Tomizuka, Mingyu Ding, Ping Luo

专题命中 机器人学习 ：manipulation(abstract)；robotic(abstract)；分类 cs.RO、cs.CV、cs.LG

Comments Accepted by CVPR 2024. Camera ready version. Project page: https://skilldiffuser.github.io/

SLIM: Skill Learning with Multiple Critics

David Emukpere, Bingbing Wu, Julien Perez, Jean-Michel Renders

专题命中 机器人学习 ：manipulation(abstract)；robotic(abstract)；分类 cs.RO、cs.AI、cs.LG

Comments Accepted at IEEE ICRA 2024

DNAct: Diffusion Guided Multi-Task 3D Policy Learning

Ge Yan, Yueh-Hua Wu, Xiaolong Wang

专题命中 机器人学习 ：manipulation(abstract)；robotic(abstract)；分类 cs.RO、cs.AI、cs.CV

Learning with Language-Guided State Abstractions

Andi Peng, Ilia Sucholutsky, Belinda Z. Li, Theodore R. Sumers, Thomas L. Griffiths, Jacob Andreas, Julie A. Shah

专题命中 机器人学习 ：manipulation(abstract)；robotic(abstract)；分类 cs.RO、cs.AI、cs.LG

Comments ICLR 2024

Hierarchical Empowerment: Towards Tractable Empowerment-Based Skill Learning

Andrew Levy, Sreehari Rammohan, Alessandro Allievi, Scott Niekum, George Konidaris

专题命中 机器人学习 ：robotics(abstract)；navigation(abstract)；分类 cs.RO、cs.AI、cs.LG

Comments Additional baseline comparisons

Hierarchical Few-Shot Imitation with Skill Transition Models

Kourosh Hakhamaneshi, Ruihan Zhao, Albert Zhan, Pieter Abbeel, Michael Laskin

专题命中 机器人学习 ：navigation(abstract)；robotic(abstract)；分类 cs.RO、cs.AI、cs.LG

Implicit Behavioral Cloning

Pete Florence, Corey Lynch, Andy Zeng, Oscar Ramirez, Ayzaan Wahid, Laura Downs, Adrian Wong, Johnny Lee, Igor Mordatch, Jonathan Tompson

专题命中 机器人学习 ：robot policy(abstract)；robotic(abstract)；分类 cs.RO、cs.CV、cs.LG

Learning from Demonstration with Weakly Supervised Disentanglement

Yordan Hristov, Subramanian Ramamoorthy

专题命中 机器人学习 ：manipulation(abstract)；robotic(abstract)；分类 cs.RO、cs.CV、cs.LG

Comments 18 pages, 16 figures, accepted at the International Conference on Learning Representations (ICLR) 2021, supplementary website at https://sites.google.com/view/weak-label-lfd

Human-Machine Interface for Remote Training of Robot Tasks

Jordi Spranger, Roxana Buzatoiu, Athanasios Polydoros, Lazaros Nalpantidis, Evangelos Boukas

专题命中 机器人学习 ：robotics(abstract)；robot learning(abstract)；分类 cs.RO、cs.CV、cs.LG

Comments Accepted in IEEE International Conference on Imaging Systems and Techniques - IST2018

Neural Task Programming: Learning to Generalize Across Hierarchical Tasks

Danfei Xu, Suraj Nair, Yuke Zhu, Julian Gao, Animesh Garg, Li Fei-Fei, Silvio Savarese

专题命中 机器人学习 ：robot learning(abstract)；manipulation(abstract)；分类 cs.RO、cs.AI、cs.LG

Comments ICRA 2018

Multi-task Learning with Gradient Guided Policy Specialization

Wenhao Yu, C. Karen Liu, Greg Turk

专题命中 机器人学习 ：manipulation(abstract)；robotic(abstract)；分类 cs.RO、cs.AI、cs.LG

Supervision via Competition: Robot Adversaries for Learning Tasks

Lerrel Pinto, James Davidson, Abhinav Gupta

专题命中 机器人学习 ：robotics(abstract)；robot learning(abstract)；分类 cs.RO、cs.CV、cs.LG

Comments Submission to ICRA 2017

Emerging Extrinsic Dexterity in Cluttered Scenes via Dynamics-aware Policy Learning

杂乱场景中通过动力学感知策略学习涌现的外在灵巧性

Yixin Zheng, Jiangran Lyu, Yifan Zhang, Jiayi Chen, Mi Yan, Yuntian Deng, Xuesong Shi, Xiaoguang Zhao, Yizhou Wang, Zhizheng Zhang, He Wang

机构 * Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所） ； Beijing Academy of Artificial Intelligence（北京人工智能研究院） ； Galbot ； Peking University（北京大学） ； Shanghai Jiao Tong University（上海交通大学） 纠错

专题命中 机器人学习 ：manipulation(abstract)；world model(abstract)；分类 cs.RO、cs.AI；robotics(comments)

AI总结 提出动力学感知策略学习框架，通过显式世界建模学习接触诱导物体动力学表示并用于强化学习，使杂乱场景中的外在灵巧性无需手工启发式或复杂奖励塑造即可涌现。

Comments Accepted to Robotics: Science and Systems (RSS) 2026. Project page: https://pku-epic.github.io/DAPL/

ReSeFlow: Rectifying SE(3)-Equivariant Policy Learning Flows

Zhitao Wang, Yanke Wang, Jiangtao Wen, Roberto Horowitz, Yuxing Han

机构 * Tsinghua University, Shenzhen International Graduate School, Shenzhen, China（清华大学深圳国际研究生院，深圳，中国） ； College of Professional and Continuing Education (CPCE), The Hong Kong Polytechnic University (PolyU)（香港理工大学（PolyU）专业及继续教育学院） ； NYU Shanghai, China（纽约大学上海学院，中国） ； Department of Mechanical Engineering, University of California, Berkeley, USA（加州大学伯克利分校机械工程系） 纠错

专题命中 机器人学习 ：manipulation(abstract)；robotic(abstract)；分类 cs.RO、cs.CV；robotics(comments)

Comments This work was submitted to 2026 IEEE International Conference on Robotics & Automation

Optimal Interactive Learning on the Job via Facility Location Planning

Shivam Vats, Michelle Zhao, Patrick Callaghan, Mingxi Jia, Maxim Likhachev, Oliver Kroemer, George Konidaris

机构 * Brown University（布朗大学） ； Carnegie Mellon University（卡内基梅隆大学） 纠错

专题命中 机器人学习 ：robot learning(abstract)；manipulation(abstract)；分类 cs.RO、cs.AI；robotics(comments)

Comments Accepted to Robotics: Science and Systems (RSS) 2025

Vid2Robot: End-to-end Video-conditioned Policy Learning with Cross-Attention Transformers

Vidhi Jain, Maria Attarian, Nikhil J Joshi, Ayzaan Wahid, Danny Driess, Quan Vuong, Pannag R Sanketi, Pierre Sermanet, Stefan Welker, Christine Chan, Igor Gilitschenski, Yonatan Bisk, Debidatta Dwibedi

专题命中 机器人学习 ：manipulation(abstract)；robotic(abstract)；分类 cs.RO、cs.AI；robotics(comments)

Comments Robotics: Science & Systems (RSS) 2024. https://vid2robot.github.io/

Time Reversal as Self-Supervision

Suraj Nair, Mohammad Babaeizadeh, Chelsea Finn, Sergey Levine, Vikash Kumar

专题命中 机器人学习 ：robot learning(abstract)；manipulation(abstract)；分类 cs.RO、cs.AI；robotics(journal_ref)

Comments 7 pages, 10 figures

Journal ref International Conference on Robotics and Automation, 2020

See, Learn, Assist: Safe and Self-Paced Robotic Rehabilitation via Video-Based Learning from Demonstration

看见、学习、辅助：基于视频学习的机器人康复安全自适应方法

Ali Alabbas, Camillo Murgia, Joanne Regan, Philip Long

专题命中 机器人学习 ：robotic(title)；分类 cs.RO

AI总结 本文提出一种基于视频的远程机器人康复教学框架，通过动态运动基元编码示范为6自由度身体中心轨迹，结合空间合规虚拟隧道和努力基 temporal dilation 机制，实现安全自适应康复训练。

Continual Robot Policy Learning via Variational Neural Dynamics

通过变分神经动力学的持续机器人策略学习

Jiaxu Xing, Zhiyuan Zhu, Yunfan Ren, Ismail Geles, Yifan Zhai, Rudolf Reiter, Davide Scaramuzza

机构 * Robotics and Perception Group, University of Zurich（苏黎世大学机器人感知组） 纠错

专题命中 机器人学习 ：robot policy(title)；分类 cs.RO

AI总结 提出持续学习框架，结合物理先验与神经残差建模隐藏动态，通过可微仿真和在线条件推断提升策略在未知干扰下的性能。