Recently, 3D vision-based diffusion policies have shown strong capability in learning complex robotic manipulation skills. However, a common architectural mismatch exists in these models: a tiny yet efficient point-cloud encoder is often paired with a massive decoder. Given a compact scene representation, we argue that this may lead to substantial parameter waste in the decoder. Motivated by this observation, we propose PocketDP3, a pocket-scale 3D diffusion policy that replaces the heavy conditional U-Net decoder used in prior methods with a lightweight Diffusion Mixer (DiM) built on MLP-Mixer blocks. This architecture enables efficient fusion across temporal and channel dimensions, significantly reducing model size. Notably, without any additional consistency distillation techniques, our method supports two-step inference without sacrificing performance, improving practicality for real-time deployment. Across three simulation benchmarks--RoboTwin2.0, Adroit, and MetaWorld--PocketDP3 achieves state-of-the-art performance with fewer than 1% of the parameters of prior methods, while also accelerating inference. Real-world experiments further demonstrate the practicality and transferability of our method in real-world settings. Code will be released.
DINOv3-Diffusion Policy: Self-Supervised Large Visual Model for Visuomotor Diffusion Policy Learning
Department of Computing and Mathematics, Manchester Metropolitan University（计算与数学系，曼彻斯特 Metropolitan 大学）；Department of Computing, Imperial College London（计算系，帝国理工学院伦敦分校）
AimBot: A Simple Auxiliary Visual Cue to Enhance Spatial Awareness of Visuomotor Policies
Computer Science and Engineering Department, University of Michigan（密歇根大学计算机科学与工程系）；Robotics Department, University of Michigan（密歇根大学机器人学系）；Lambda Labs（Lambda实验室）
FreqPolicy: Frequency Autoregressive Visuomotor Policy with Continuous Tokens
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）
If a robot masters folding a kitchen towel, we would expect it to master folding a large beach towel. However, existing policy learning methods that rely on data augmentation still don't guarantee such generalization. Our insight is to add equivariance to both the visual object representation and policy architecture. We propose EquivAct which utilizes SIM(3)-equivariant network structures that guarantee generalization across all possible object translations, 3D rotations, and scales by construction. EquivAct is trained in two phases. We first pre-train a SIM(3)-equivariant visual representatio…
Scaling Diffusion Policy in Transformer to 1 Billion Parameters for Robotic Manipulation
School of Computer Science, East China Normal University（华东师范大学计算机科学学院）；Midea Group, AI Research Center（美的集团人工智能研究院）；Department of Computer Science, Shanghai University（上海大学计算机科学系）；Beijing Innovation Center of Humanoid Robotics（北京人形机器人创新中心）
CordViP: Correspondence-based Visuomotor Policy for Dexterous Manipulation in Real-World
State Key Laboratory of Multimedia Information Processing, School of Computer Science, Peking University（多媒体信息处理国家重点实验室，计算机学院，北京大学）；The University of Hong Kong（香港大学）；Beijing Academy of Artificial Intelligence（北京人工智能研究院）
Risk-Aware Reinforcement Learning for Mobile Manipulation
Oxford Robotics Institute, Department of Engineering Science, University of Oxford（牛津大学机器人研究所、工程科学系、牛津大学）；Dyson Institute of Engineering & Technology（戴森工程与技术研究所）；Bristol Robotics Laboratory, School of Engineering, University of West England（布里斯托尔机器人实验室、工程学院、西英格兰大学）
KineDex: Learning Tactile-Informed Visuomotor Policies via Kinesthetic Teaching for Dexterous Manipulation
Tongji University（同济大学）；Tsinghua University（清华大学）；Shanghai Jiao Tong University（上海交通大学）；University of Hong Kong（香港大学）；Shanghai Qi Zhi Institute（上海启智研究院）；Shanghai AI Lab（上海人工智能实验室）
Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所）；School of Artificial Intelligence, University of Chinese Academy of Sciences（中国科学院大学人工智能学院）；State Key Laboratory of Brain Cognition and Brain-Inspired Intelligence Technology（脑认知与脑启发智能技术国家重点实验室）；Institute of Semiconductors, Chinese Academy of Sciences（中国科学院半导体研究所）
Learning Adaptive Cross-Embodiment Visuomotor Policy with Contrastive Prompt Orchestration
School of Mechanical and Aerospace Engineering, Nanyang Technological University（南洋理工大学机械与航空航天工程学院）；College of Automation Engineering, Nanjing University of Aeronautics and Astronautics（南京航空航天大学自动化工程学院）
CUBic: Coordinated Unified Bimanual Perception and Control Framework
Beijing Advanced Innovation Center for Future Blockchain and Privacy Computing, School of Artificial Intelligence, Beihang University（北京未来区块链与隐私计算高级创新中心，人工智能学院，北京航空航天大学）；Westlake University（西湖大学）；Zhejiang University（浙江大学）；Peking University（北京大学）
A Study on Enhancing the Generalization Ability of Visuomotor Policies via Data Augmentation
University of Chinese Academy of Sciences（中国科学院大学）；School of Artificial Intelligence & Institute of Automation, Chinese Academy of Sciences & NLPR（中国科学院人工智能学院与自动化研究所）
In this paper we explore using self-supervised correspondence for improving the generalization performance and sample efficiency of visuomotor policy learning. Prior work has primarily used approaches such as autoencoding, pose-based losses, and end-to-end policy optimization in order to train the visual portion of visuomotor policies. We instead propose an approach using self-supervised dense visual correspondence training, and show this enables visuomotor policy learning with surprisingly high generalization performance with modest amounts of data: using imitation learning, we demonstrate ex…
KTH Royal Institute of Technology（瑞典皇家理工学院）；Aalto University（阿尔托大学）
We present a data-efficient framework for solving visuomotor sequential decision-making problems which exploits the combination of reinforcement learning (RL) and latent variable generative models. Our framework trains deep visuomotor policies by introducing an action latent variable such that the feed-forward policy search can be divided into three parts: (i) training a sub-policy that outputs a distribution over the action latent variable given a state of the system, (ii) unsupervised training of a generative model that outputs a sequence of motor actions conditioned on the latent action var…
