Huaxiu Yao

My research focuses on both the theoretical and applied aspects of building generalizable and adaptive agentic foundation models. Additionally, I am keen on utilizing these models to facilitate applications in biomedicine/healthcare, education, and robotics. Currently, my primary endeavors revolve around the following key directions: 1. Enabling continual adaptability and evolvability of AI agents in unseen environments and tasks. 2. Exploring effective strategies to fine-tune foundation agents for enhanced generalization, reasoning, and alignment. 3. Pioneering interactive systems that seamlessly integrate humans, foundation models, and skills to achieve complex goals.

[1] 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*, MetaClaw: Just Talk – An Agent That Meta-Learns and Evolves in the Wild, arXiv 2603.17187. [2] Peng Xia*, Jianwen Chen*, Hanyang Wang*, Jiaqi Liu, Kaide Zeng, Yu Wang, Siwei Han, Yiyang Zhou, Xujiang Zhao, Haifeng Chen, Zeyu Zheng, Cihang Xie, Huaxiu Yao, SkillRL: Evolving Agents via Recursive Skill-Augmented Reinforcement Learning, arXiv 2602.08234. ICLR 2026 MemAgents Best Paper Award Runner-Up [3] Jiaqi Liu, Zipeng Ling, Shi Qiu, Yanqing Liu, Siwei Han, Peng Xia, Haoqin Tu, Zeyu Zheng, Cihang Xie, Charles Fleming, Mingyu Ding, Huaxiu Yao, Omni-SimpleMem: Autoresearch-Guided Discovery of Lifelong Multimodal Agent Memory, arXiv