Peng “Richard” Xia夏鹏

I am a CS Ph.D. candidate at The University of North Carolina at Chapel Hill, advised by Prof. Huaxiu Yao. I am also a student researcher at Google, working on recursive self-improvement. Previously, I conducted research at Tongyi Lab, Alibaba and Microsoft Research.

I build long-horizon LLM & VLM agents through mid/post-training (imitation or reinforcement learning) and data/environment scaling. My research focuses on equipping recursively self-improving agents with skills/memory and complex tool-use capabilities to tackle open-ended workflows that require several human hours.

My work has been accepted as oral or spotlight presentations at ICLR (2025, 2026), NeurIPS (2025), ICML (2026) and ACL (2026), and received the Best Paper Runner-Up Award at the ICLR MemAgents Workshop and the Outstanding Paper Award at the ICLR RSI Workshop.