Adaptive Action Chunking at Inference-time for Vision-Language-Action Models
National University of Singapore(新加坡国立大学);Shenzhen University of Advanced Technology(深圳理工大学);Sangfor Technologies Inc.(深信服科技股份有限公司);Mininglamp Technology(明略科技);Shenzhen Technology University(深圳技术大学);City University of Hong Kong(香港城市大学)
Yuanchang Liang, Xiaobo Wang, Kai Wang, Shuo Wang, Xiaojiang Peng, Haoyu Chen, David Kim Huat Chua, Prahlad Vadakkepat
In Vision-Language-Action (VLA) models, action chunking (i.e., executing a sequence of actions without intermediate replanning) is a key technique to improve robotic manipulation abilities. However, a large chunk size reduces the model's responsiveness to new information, while a small one increases the likelihood of mode-jumping, jerky behavior resulting from discontinuities between chunks. Therefore, selecting the optimal chunk size is an urgent demand to balance the model's reactivity and consistency. Unfortunately, a dominant trend in current VLA models is an empirical fixed chunk length at inference-time, hindering their superiority and scalability across diverse manipulation tasks. To address this issue, we propose a novel Adaptive Action Chunking (AAC) strategy, which exploits action entropy as the cue to adaptively determine the chunk size based on current predictions. Extensive experiments on a wide range of simulated and real-world robotic manipulation tasks have demonstrated that our approach substantially improves performance over the state-of-the-art alternatives. The videos and source code are publicly available at https://lance-lot.github.io/adaptive-chunking.github.io/.
Commit While Futures Agree: Consequence-Aware Adaptive Action Chunking for Robot Manipulation
The Hong Kong University of Science and Technology (Guangzhou)(香港科技大学(广州));ANU Intelligent(澳大利亚国立大学智能);Victoria University of Wellington(惠灵顿维多利亚大学);The Hong Kong University of Science and Technology(香港科技大学);Southern University of Science and Technology(南方科技大学)
Memorized action chunking with Transformers: Imitation learning for vision-based tissue surface scanning
Imperial College London(帝国理工学院);Hamlyn Centre for Robotic Surgery(哈姆林机器人外科中心);Institute of Global Health Innovation(全球健康创新研究所);Department of Surgery and Cancer(外科与癌症系)
Denoising Tells When to Replan: Denoising-Variance Adaptive Chunking for Flow-Based Robot Policies
Beijing Institute of Technology(北京理工大学);The Chinese University of Hong Kong, Shenzhen(香港中文大学(深圳));Shenzhen Loop Area Institute(深圳河套学院);Hunan University(湖南大学);Xi’an Jiaotong University(西安交通大学);Renmin University of China(中国人民大学);Harbin Institute of Technology, Shenzhen(哈尔滨工业大学(深圳))
Colosseum V2: Benchmarking Generalization for Vision-Language-Action Models
Department of Computer Science, University of Southern California(南加州大学计算机科学系);Department of Electrical Engineering, Indian Institute of Technology Madras(印度理工学院Madras分校电子工程系);Fu Foundation School of Engineering and Applied Science, Columbia University(哥伦比亚大学工程与应用科学学院)
Enabling Dynamic Tracking in Vision-Language-Action Models via Time-Discrete and Time-Continuous Velocity Feedforward
Siemens Foundational Technologies(西门子基础技术);Department of Computer Science & Artificial Intelligence, University of Technology Nuremberg(技术大学纽伦堡计算机科学与人工智能系)
ATAAT: Adaptive Threat-Aware Adversarial Tuning Framework against Backdoor Attacks on Vision-Language-Action Models
Chongqing Institute of Green and Intelligent Technology, Chinese Academy of Sciences(重庆绿色智能技术研究院,中国科学院);Chongqing School, University of Chinese Academy of Sciences(重庆学校,中国科学院大学);Faculty of Information Technology and Electrical Engineering, University of Oulu, Finland(奥卢大学信息科技与电气工程学院,芬兰)
HiPolicy: Hierarchical Multi-Frequency Action Chunking for Policy Learning
CFCS, School of CS, PKU(北京大学计算机学院计算与金融研究中心);National Key Laboratory for Multimedia Information Processing, School of CS, PKU(北京大学计算机学院多媒体信息处理国家重点实验室);XJTU(西安交通大学);THU(清华大学);BUAA(北京航空航天大学);Jingdong Technology Information Technology Co., Ltd(京东科技信息技术有限公司)
Cross-Modal Visuo-Tactile Representation Learning with Action Chunking Transformers for Contact-Rich Manipulation
Guangdong Institute of Intelligence Science and Technology(广东智能科学与技术研究院);Innovation and Research and Development Department, BoardWare Information System Company Ltd.(创新与研发部,BoardWare信息系统有限公司);School of Artificial Intelligence, Nanjing Agricultural University(人工智能学院,南京农业大学);School of Computing, National University of Singapore(计算机学院,新加坡国立大学);School of Electronic Engineering and Automation, Guilin University of Electronic Technology(电子工程与自动化学院,桂林电子科技大学)
Implicit Action Chunking for Smooth Continuous Control
Department of Data and Systems Engineering, The University of Hong Kong, Hong Kong SAR, China(香港大学数据与系统工程系);Beijing Institute of Technology, Zhuhai, China(北京理工大学珠海学院);College of Computer Science, Sichuan University, Chengdu, China(四川大学计算机学院)
Bi-LAT: Bilateral Control-Based Imitation Learning via Natural Language and Action Chunking with Transformers
Graduate School of Information Science and Technology, The University of Osaka(信息科学与技术研究生学校,大阪大学);D3 Center, The University of Osaka(大阪大学D3中心);Graduate School of Maritime Sciences, Kobe University(海洋科学研究生学校, Kobe大学)
Learning When to See and When to Feel: Adaptive Vision-Torque Fusion for Contact-Aware Manipulation
Edwardson School of Industrial Engineering, Purdue University(普渡大学爱德华森工业工程学院);Zachry Department of Civil and Environmental Engineering, Texas A&M University(德州农工大学扎克里土木与环境工程系)
From Vision to Assistance: Gaze and Vision-Enabled Adaptive Control for a Back-Support Exoskeleton
Department of Innovative Technologies (DTI), Dalle Molle Institute for Artificial Intelligence (IDSIA), University of Applied Sciences and Arts of Southern Switzerland (SUPSI)(创新技术系(DTI)、人工智能研究所(IDSIA)、瑞士应用科学与艺术大学(SUPSI));Department of Informatics, Systems and Communication (DISCo), University of Milano-Bicocca(信息、系统与通信系(DISCo)、米兰-比科卡大学);Department of Mechanical Engineering, Politecnico di Milano(机械工程系、米兰理工学院)