arXiv Robotics — research abstracts· Changchuan Yang, Haoxuan Xu, Wenbo Chen, Shuai Ren, Jianlong Zheng, Huarui Zhang, Tianfu Li, Guanzhong Tian·· 10 days agoEditorial score43
PMTRM: A Lightweight Module for Embodied Policy Learning
PMTRM: Pseudo-Memory Temporal Re-encoding Module for Embodied Policy Learning
Summary
PMTRM improves task success in phase-ambiguous scenarios by encoding historical states and actions. It uses temporal heterogeneity and reconstruction losses to enhance policy learning without altering the original action space.
Source: arXiv Robotics — research abstracts · Read original article ↗
Loading article text…
Source：arXiv Robotics — research abstracts · arxiv.org