In-context robot learning enables robots to acquire new behaviors from demonstrations without test-time parameter updates. However, visual and kinematic specifications alone may not distinguish contact-rich behaviors that share similar motions but require different interaction intensities or patterns. We propose Contac...
Xiang-Li Nie, Yujie Zang, Yu-Hang Zheng et al.· 0 citations
MAESTRO is used as a coding agent that writes robot programs using a curated library of perception, planning, and control tools and its orchestration harness packages multi-step perception and action procedures into callable tools and guides their use through a plan–act– replan loop.
Jun-Yao Shi, Ru-Jia Yang, Kai-Tian Chao et al.· 0 citations
Real-world post-training of generalist robot policies faces two challenges: partial observability makes action selection ambiguous, while offline post-training methods delay feedback from deployment experience. We present Value as Memory (VaM), a framework that addresses these challenges through value caching and onlin...
Wen-Ye Yu, Yi-Jin Chen, Kai Xiong et al.· 0 citations
A generalist robot should not only perform diverse tasks but also improve through experience, turning what it learns during execution into capabilities that later tasks can reuse. Robot agents that act through code can already repair programs from execution feedback, yet it remains a central challenge to organize this...
Zi-Mo Wen, Yi-Jin Chen, Yuxuan Cao et al.· 8 citations· ⚡1
A Unified Visual-Tactile-Action Model is proposed that maps both embodiments into aligned tactile and action representations and jointly predicts future action and tactile trajectories and achieves an average success rate of 70%, outperforming the strongest visual-tactile baseline, which achieves 29%, and an architectu...
Wen-Qiao Li, Qian-You Zhao, Jia-Wen Hao et al.· 0 citations
Across towel folding, book insertion, and Hanoi ring placement, LIFT learns faster and reaches higher performance than vision-only post-training, while ablations show that reactive force memory and online corrective data are both important for robust contact-rich manipulation.
Yi Wang, Wen-Di Chen, Zi-Mo Wen et al.· arXiv.org· 4 citations
A system level acceleration strategy that reduces computation in both perception and action generation and compress diffusion sampling into a compact 2-step schedule through efficiency oriented training while preserving action precision is proposed.
The Triplet-to-Track System (TTS), a closed-loop long-horizon imitation learning system that uses human videos to reduce reliance on robot-collected data, achieves a 74.8\% average success rate and supports object-level and compositional generalization.
Jianxiang Liu, Gao-Jing Zhang, Chuan Wen et al.· 0 citations
Results show that a fixed-weight, self-evolving harness can revise, recover, and accumulate verified approaches while producing structured trajectories for future supervised and reinforcement learning.
Boxiu Li, Zi-Mo Wen, Yi-Jia Fan et al.· 2 citations· ⚡1
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
