Robots Pre-train Robots: Manipulation-Centric Robotic Representation from Large-Scale Robot Datasets
Guangqi Jiang, Yifei Sun, Tao Huang, Huanyu Li, Yongyuan Liang, Huazhe Xu
Abstract
DexArt MCR (ours) R3M R3M-DROID MVP VC-1 HRP Figure 2 : Correlation between manipulation centricity and downstream performance. Our findings reveal that (1) the proposed metric of manipulation centricity strongly correlates with the downstream performance of robotic representations, and (2) using the robot dataset DROID yields greater benefits for robotic representations than human datasets. (3) These insights motivate our method, MCR, which leverages dynamics labels from the robot dataset to further enhance manipulation centricity and downstream performance.
Ask about this paper
Your agent reads all of it.
Lune indexed this paper to the last equation, along with the top-tier papers that cite it. Ask a question and the answer quotes them.
Cited by top-tier papers11
- TraceGen: World Modeling in 3D Trace Space Enables Learning from Cross-Embodiment VideosSeungjae Lee, Yoonkyo Jung, Inkook Chun, Yao-Chih Lee et al.CVPR 2026 · 17 citations
- Cross-Hand Latent Representation for Vision-Language-Action ModelsGuangqi Jiang, Yutong Liang, Jianglong Ye, Jia-Yang Huang et al.CVPR 2026 · 14 citations
- When would Vision-Proprioception Policies Fail in Robotic Manipulation?Jingxian Lu, Wenke Xia, Yuxuan Wu, Zhiwu Lu et al.ICLR 2026 · 11 citations
- Bootstrap Dynamic-Aware 3D Visual Representation for Scalable Robot LearningQiwei Liang, Boyang Cai, Minghao Lai, Sitong Zhuang et al.CVPR 2026 · 6 citations
- Contact-Aware Neural DynamicsChangwei Jing, Jai Krishna Bandi, Jianglong Ye, Yan Duan et al.CVPR 2026 · 5 citations
Builds on17
- An Image is Worth 16x16 Words: Transformers for Image Recognition at ScaleAlexey Dosovitskiy, Lucas Beyer, Alexander Kolesnikov, Dirk Weissenborn et al.ICLR 2021 · 21,477 citations
- Data-Efficient Image Recognition with Contrastive Predictive CodingOlivier J. HénaffICML 2020 · 1,553 citations
- CURL: Contrastive Unsupervised Representations for Reinforcement LearningMichael Laskin, Aravind Srinivas, Pieter AbbeelICML 2020 · 1,261 citations
- Masked Autoencoders As Spatiotemporal LearnersChristoph Feichtenhofer, Haoqi Fan, Yanghao Li, Kaiming HeNeurIPS 2022 · 690 citations
- Decoupling Representation Learning from Reinforcement LearningAdam Stooke, Kimin Lee, Pieter Abbeel, Michael LaskinICML 2021 · 389 citations
Related papers
- Learning Object-Centric Motion Priors from Human for Robotic Dexterous ManipulationZhengdong Hong, Guofeng ZhangAAAI 2026
- EC-Flow: Enabling Versatile Robotic Manipulation from Action-Unlabeled Videos via Embodiment-Centric FlowYixiang Chen, Peiyan Li, Yan Huang, Jiabing Yang et al.ICCV 2025 · 2 citations
- Hand-Centric Motion Refinement for 3D Hand-Object Interaction via Hierarchical Spatial-Temporal ModelingYuze Hao, Jianrong Zhang, Tao Zhuo, Fuan Wen et al.AAAI 2024 · 7 citations
- HiMaCon: Discovering Hierarchical Manipulation Concepts from Unlabeled Multi-Modal DataRuizhe Liu, Pei Zhou, Qian Luo, Li Sun et al.NeurIPS 2025 · 2 citations
- Human2Robot: Learning Robot Actions from Paired Human-Robot VideosSicheng Xie, Haidong Cao, Zejia Weng, Zhen Xing et al.AAAI 2026 · 15 citations
