About Me

I work at ByteDance Seed, as a senior research scientist and a member of TopSeed program. I am working on advancing the reasoning and agentic capabilities of Large Language Models (LLMs) and general Agent foundation models. As one of the algorithm leads for general agent optimization, I contribute to the ByteDance Seed model series (e.g., Seed 2.1 to Seed 1.5) and Doubao products, including building the Agent Task Mode/Expert Mode (豆包办公任务模式和专家模式). We are hiring research interns and looking for academic cooperation, please feel free to email me at wanjun@bytedance.com

Prior to that, I worked at Huawei Noah’s Ark Lab as a research scientist and a member of Huawei TopMind program. I received my Ph.D. degree from the School of Computer Science and Engineering in Sun Yat-sen University (SYSU), as a member of joint Ph.D. program between SYSU and Microsoft Research Asia (MSRA).

As a joint Ph.D. student, I was advised by Dr. Ming Zhou, Prof. Jian Yin and Prof. Jiahai Wang. I was a research intern in the Natural Language Computing Group of MSRA, and was mentored by Dr. Nan Duan

I won the Microsoft Research Fellowship Award (11 outstanding Ph.D. in Asia-Pacific area each year) in 2021, and is selected as a member of Huawei TopMind program in 2023.

I published over 50+ papers in top-tier AI conferences and journals, including NeurIPS, ICLR, ACL, EMNLP, TASLP, NAACL, AAAI, IJCAI, ISSTA, etc.

My Research Interests

• Large Language Model
• Agent Foundation Model and Reinforcement Learning
• Reasoning towards AGI