I am a senior researcher at Tencent AI Lab, specializing in Search and Reinforcement Learning (RL) to enhance the reasoning capabilities of large language models (LLMs). My research focuses on improving the scaling efficiency of both Search and Learning processes. I have been working in the field of Natural Language Processing (NLP) since 2010. I hold a Ph.D. in Computer Science from the University of Rochester (UofR), where I was advised by Professor Daniel Gildea, and a master’s degree from the Institute of Computing Technology, Chinese Academy of Sciences (ICT/CAS), under the mentorship of Dr. Qun Liu.

Recent News

• 2024.11 Checkout our recent work on improving the scaling efficiency of search and learning: AlphaLLM (Neurips 2024), LiteSearch (AAAI 2025), StepwiseCPL and HunyuanProver.
• 2024.02 I will serve as an Action Editor of ACL ARR in 2024.
• 2023.09 Checkout our recent papers on industrial-level post-training practices: Stabilizing RLHF, Collaborative Decoding and Reward Stability in RLHF (ICLR 2024).
• 2022.04 I’m honored to serve as a Senior Area Chair at EMNLP 2022.
• 2021.05 I’m so excited that our paper “Video-aided Unsupervised Grammar Induction” was selected as the Best Long Paper at the 2021 Annual Conference of the North American Chapter of the Association for Computational Linguistics (NAACL).
• 2020.12 I will serve as a senior PC member (≈Area Chair) at IJCAI 2021.
• 2020.11 I will join the editorial team of TACL, a top journel focusing on NLP.
• 2020.09 I will give a talk about the current progress on AMR parsing and AMR-to-text generation at CCL 2020.
• 2020.03 I will serve as an Area Chair for the “Semantics” track of EMNLP 2020.