I am a researcher at Qwen Team, Token Foundry, where I contribute to the training of Qwen-series frontier LLMs. I received my PhD degree from the National University of Singapore (NUS), advised by Prof. Bryan Hooi and funded by the President's Graduate Fellowship. My research aims to build agentic AI systems that can improve themselves, learning continuously from interaction, feedback, and experience. To support this goal, I extensively work on post-training for LLMs, to strengthen their reasoning and agentic behaviors.
I have published over 20 papers at top-tier (CCF-A) AI/ML conferences and journals, including ICLR, ICML, NeurIPS, and IEEE TPAMI, where I also serve as an Area Chair or reviewer.
Previously, I worked at Microsoft Research Singapore, ByteDance, Tencent, and Z.ai (Zhipu AI).
For a full and up-to-date list of my publications, please see my Google Scholar page.