Robot-DIFT: Correspondence-Sensitive Diffusion Features for Contact-Rich Robot Manipulation
TU Darmstadt(图宾根大学);KIT(卡尔斯鲁厄理工学院);FZI(弗劳恩霍夫研究所);Hessian.AI(黑森人工智能公司);Robotics Institute Germany(德国机器人研究所);Honda Research Institute Europe GmbH(本田欧洲研究院)
Robot manipulation often fails in the final millimeters: a policy may recognize the right object yet miss the pose offsets, boundaries, or pre-contact alignments needed for action. We argue that such failures arise when semantic invariance suppresses correspondence cues for closed-loop control, or when these cues are not exposed to the policy in a usable form. Modern visual encoders provide strong semantic abstractions, but contact-rich manipulation requires correspondence sensitivity: discriminative feature responses to action-relevant changes in pose, boundary, and contact geometry. Diffusion features provide a strong prior for dense correspondence, but direct use is impractical due to stochasticity, latency, and representation drift. We introduce Robot-DIFT, a deterministic diffusion-derived backbone for real-time control. Through Manifold Distillation, Robot-DIFT converts a noise-conditioned diffusion Teacher into a clean-input, single-pass Student while preserving the teacher's feature manifold. A Spatial--Semantic Feature Pyramid Network (S2-FPN) fuses coarse-to-fine Student decoder features into visual tokens that expose semantic context and fine contact detail to the policy. Across RoboCasa, LIBERO-10, and real robots, Robot-DIFT outperforms vision--language, self-supervised, geometry-oriented, and diffusion baselines on contact-sensitive tasks. Controlled backbone/readout swaps show that S2-FPN unlocks, rather than replaces, the diffusion correspondence prior.
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)
GEM-4D: Geometry-Enhanced Video World Models for Robot Manipulation
Harvard AI and Robotics Lab(哈佛人工智能与机器人实验室);Harvard University(哈佛大学);Media Lab and EECS(媒体实验室和电子工程与计算机科学系);MIT(麻省理工学院);Princeton University(普林斯顿大学);MIT-IBM Watson AI Lab(麻省理工-IBM沃森人工智能实验室)
InCoM: Intent-Driven Perception and Structured Coordination for Mobile Manipulation
Institute of Automation, Chinese Academy of Sciences(中国科学院自动化研究所);School of Advanced Interdisciplinary Sciences, University of Chinese Academy of Sciences(中国科学院大学先进交叉学科学院);AnyverseDynamics
SemAnCorr: Semantic Anchored Correspondence for Zero-Shot Manipulation Skill Transfer
Robotics Institute, Carnegie Mellon University(卡内基梅隆大学机器人研究所);College of Connected Computing, Vanderbilt University(范德堡大学互联计算学院);College of Engineering, University of California, Berkeley(加州大学伯克利分校工程学院);School of Computer Science, The University of Sydney(悉尼大学计算机学院)
SparseDFF: Sparse-View Feature Distillation for One-Shot Dexterous Manipulation
CFCS, School of Computer Science, Peking University(北京大学计算机学院CFCS);Institute for AI, Peking University(北京大学人工智能研究院);Department of Computer Science, Stanford University(斯坦福大学计算机科学系);PKU-WUHAN Institute for Artificial Intelligence, China(北京大学武汉人工智能研究院)
MLA: A Multisensory Language-Action Model for Multimodal Understanding and Forecasting in Robotic Manipulation
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,计算机科学学院,北京大学);Beijing Innovation Center of Humanoid Robotics (X-Humanoid)(北京类人机器人创新中心(X-Humanoid));The Chinese University of Hong Kong (CUHK)(香港中文大学(CUHK))
CLAR: Learning 3D Representations for Robotic Manipulation by Fusing Masked Reconstruction with Multi-Level Contrastive Alignment
SKL-MAIS, Institute of Automation, Chinese Academy of Sciences(中国科学院自动化研究所SKL-MAIS);School of Artificial Intelligence, University of Chinese Academy of Sciences(中国科学院大学人工智能学院);Carnegie Mellon University(卡内基梅隆大学);Galbot;CFCS, School of Computer Science, Peking University(北京大学计算机科学与技术学院CFCS)
KGN-Pro: Keypoint-Based Grasp Prediction through Probabilistic 2D-3D Correspondence Learning
Zhejiang University(浙江大学);Technical University of Munich(慕尼黑技术大学);Munich Center for Machine Learning(慕尼黑机器学习中心);The State Key Laboratory of Industrial Control Technology(工业控制技术国家重点实验室);China Research and Development Academy of Machinery Equipment(中国机械装备研究开发院)
Learning Reward Functions for Robotic Manipulation by Observing Humans
CNRS(法国国家科学研究中心);Inria(法国国家信息与自动化研究所);Univ. Grenoble Alpes(格勒诺布尔阿尔卑斯大学);Grenoble INP(格勒诺布尔国立理工学院);Google Research(谷歌研究院);Center for Data Science, New York University(纽约大学数据科学中心)
LaST-HD: Learning Latent Physical Reasoning from Scalable Human Data for Robot Manipulation
State Key Laboratory of Multimedia Information Processing, School of Computer Science, Peking University(北京大学计算机学院多媒体信息处理国家重点实验室);The Chinese University of Hong Kong(香港中文大学);Simplexity Robotics;Aether Tech
TacSE3: Equivariant SE(3) Motion Estimation from Low-Texture Visuotactile Images for In-Gripper Tracking and Compensation
Hong Kong Center for Construction Robotics, The Hong Kong University of Science and Technology, Hong Kong, China(香港建设机器人中心,香港科技大学,中国香港);The Hong Kong University of Science and Technology (Guangzhou), China(香港科技大学(广州),中国);School of Engineering, Great Bay University, Dongguan, China(工程学院,大湾大学,中国东莞)
SkelWAM: A Skeleton-Guided World-Action Model for Zero-Shot Cross-Embodiment Manipulation
School of Advanced Manufacturing and Robotics, Peking University(北京大学先进制造与机器人学院);Feagine;IIIS, Tsinghua University(清华大学智能产业研究院);Research Center for Robotics, Peking University(北京大学机器人研究中心)
Click to Grasp: Zero-Shot Precise Manipulation via Visual Diffusion Descriptors
School of Informatics, University of Edinburgh(爱丁堡大学信息学院);Edinburgh Centre for Robotics(爱丁堡机器人中心);Department of Computer Science, University College London(伦敦大学学院计算机科学系)