arXiv cs.AI· Python Song, Zhixuan Liang, Kelsey Fu, Mengdi Wang, Junfeng Yang, Shilong Liu·· 3 小时前AI 评分42
EmbodiedRSI：假设引导共进化的主动持续机器人学习
EmbodiedRSI: Active Continual Robot Learning Through Hypothesis-Guided Co-Evolution
AI 导读
EmbodiedRSI 是一个自进化的智能体框架，自主决定下一步探索位置，并通过假设引导的物理交互持续改进机器人的代码与技能。其采用快慢双系统架构，用假设图维护相互竞争的代码与技能假设，并通过信息价值实验选择物理实验。在 RoboCasa365 上整体成功率达 77.0%（最佳基线为 40.1%），LIBERO-Pro 上达 86.8%，并零样本迁移到真实机器人，成功率达 71.3%。
正文 · 原文
Abstract:Robot foundation models provide strong visuomotor control, yet their performance can degrade when object positions or task instructions change. Further improvements often require post-training on substantial robot data, which can be costly to collect through methods such as teleoperation. Agentic harnesses can adapt around the model, but current self-evolving harnesses use robot trials inefficiently when deciding which code and skill changes to pursue. We introduce EmbodiedRSI, a self-evolving agentic harness that autonomously decides where to explore next and turns the resulting physical interaction into improved code and skills. EmbodiedRSI realizes this through a Fast-Slow Dual-System Architecture, in which competing code and skill hypotheses are maintained in a Hypothesis Graph. Value-of-Information Experiment Selection chooses physical experiments that can distinguish these hypotheses. Their outcomes guide Code-Skill Co-Evolution. The Slow System builds Hierarchical Memory, and Reward-Grounded Memory Learning selects effective memory according to their value for later Fast-System improvement. On RoboCasa365, EmbodiedRSI reaches 77.0% overall success and 71.3% on Composite-Unseen, compared with 40.1% for the best baseline. EmbodiedRSI also reaches 86.8% overall success on LIBERO-Pro. Beyond benchmark performance, EmbodiedRSI transfers zero-shot to real-world robot, achieving 71.3% overall success across multiple challenging tasks.
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.10498 [cs.AI] |
| (or arXiv:2610.10498v2 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.10498 arXiv-issued DOI via DataCite |
Submission history
From: Python Song [view email]
[v1]
Wed, 7 Oct 2026 17:48:02 UTC (4,283 KB)
[v2]
Thu, 12 Nov 2026 04:07:56 UTC (5,228 KB)
来源：arXiv cs.AI · arxiv.org