arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2610.20474cs.CLcs.RO
收藏

AdaRSI:面向具身智能体的按需递归自我改进框架

AdaRSI: On-Demand Recursive Self-Improvement for Embodied Agents

Northeastern University(东北大学);AgiBot;University of Chinese Academy of Sciences(中国科学院大学)

Peidong Wang, Ying Chang, Zhiming Ma, Lirong Che, Zihao Liu, Xiaocui Yang, Shi Feng, Daqi Liu, Yuzhe Yang, Xu Cao, Chuang Wang, Jian Su, Daling Wang

AI总结:

针对具身智能体预训练 VLA 处理长任务易中途失败的问题,提出 AdaRSI 按需 RSI 框架,仅微调不可靠子步骤,在 RoboCasa365 任务上大幅提升成功率且资源消耗远低于全微调。

摘要:

预训练的视觉-语言-动作模型(VLA)在处理未见过的长 horizon 任务时常常中途失败。递归自我改进(RSI)提供了一种解决方案,但现有的具身 RSI 方法要么围绕冻结的执行器优化控制流程,无法添加缺失的动作;要么在持续失败时训练执行器,却未探究失败源于执行器的调用方式还是其自身能力。本文研究具身自我改进系统何时应调整执行器的权重,提出 AdaRSI——一种面向具身智能体的按需 RSI 框架。编码智能体先提升自身技能与控制流程,再通过从保存状态重试失败子步骤来归因持续失败,仅微调仍不可靠的子步骤,所用数据为其自行收集的演示及从先前任务继承的经验。在 RoboCasa365 的全部 50 个目标任务中,以 GR00T N1.5 为执行器且无人类演示的条件下,AdaRSI 在所有三个套件上均提升了 VLA 性能,将未见过的复合任务成功率从 4.6% 提升至 13.8%;与微调所有子步骤相比,它达到了略高的成功率,仅使用 28% 的 GPU 时间和 22% 的自行收集演示,而无控制流程的微调训练量更大但收益少得多。

英文摘要:

Failures in long-horizon embodied tasks can arise from how a vision-language-action model (VLA) is called or from its execution capabilities. We introduce AdaRSI, an on-demand recursive self-improvement framework in which a coding agent first refines its tools and procedures, then adapts sub-steps that remain unreliable after testing alternative instructions and approaches. The agent collects demonstrations, tests their utility through temporary fine-tuning probes, and pools accepted data across tasks to train a specialist VLA that works alongside the original executor. At the end of each development round, its records are summarized into shared experience that guides subsequent diagnosis and data collection. On unseen composite tasks in RoboCasa365, AdaRSI raises success from 4.6% to 13.3%. On perturbed LIBERO-Pro 2.0 tasks, success rises from 32.9% to 49.1%. On RoboCasa365's Composite-Unseen suite, it achieves higher success than adapting every manipulation sub-step, using 34% of the GPU time and 36% of the demonstrations. On a real robot, the agent autonomously collects demonstrations and adapts the executor to recover fallen bottles, improving task completion.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑