Skip to content

Source paper

MetaClaw: Just Talk -- An Agent That Meta-Learns and Evolves in the Wild

Peng Xia, Jianwen Chen, Xinyu Yang and 10 more

Peng Xia, Jianwen Chen, Xinyu Yang, Haoqin Tu, Jiaqi Liu, Kaiwen Xiong, Siwei Han, Shi Qiu, Haonian Ji, Yuyin Zhou, Zeyu Zheng, Cihang Xie, Huaxiu Yao

arXiv 2603.17187 · 2026102 upvotes on Hugging Face

Research ideas from this paper (3)

Generated by AI. The paper’s authors did not write these.

AI summary

The paper introduces MetaClaw, a continual meta-learning framework designed to enable deployed Large Language Model (LLM) agents to autonomously adapt to shifting task distributions in the wild. By bridging fast skill distillation and slow weight optimization, it allows agents to continuously improve from failure trajectories without requiring service downtime.

How it works (AI summary)

MetaClaw coordinates two mechanisms: a 'Skill-driven fast adaptation' that uses an LLM to synthesize reusable behavioral prompt instructions from failed interactions, and an 'Opportunistic policy optimization' that refines the base model's weights via RL (Cloud LoRA) exclusively during user-inactive windows, using a versioning system to separate pre-adaptation and post-adaptation data.