CV
A brief version of my CV is presented below. You can also download the complete version here.
Education Background
- Doctor of Philosophy, The Chinese University of Hong Kong
- August 2025 - July 2029 (expected)
- Department of Information Engineering (Multimedia Laboratory)
- Funded by the Hong Kong PhD Fellowship Scheme
- Research Interests: LLM-based Agents, Multi-agent Systems, and Agentic RL
- Bachelor of Engineering, Shanghai Jiao Tong University
- September 2021 - June 2025
- Major in Artificial Intelligence (Honor Class)
- GPA 95.44/100, Rank 1/94
- A+ Courses: Mathematical Analysis, Linear Algebra, Algorithm Design, Deep Learning, and 34 others
Selected Honors
- Hong Kong PhD Fellowship, April 2025
- Attracting top-tier students from all over the world to pursue their PhD studies in Hong Kong
- Outstanding Graduate in Shanghai, June 2025
- Awarded to top 5% undergraduate students graduating from universities across Shanghai
- National Scholarship (Three Times), December 2022, 2023, 2024
- Awarded to top 0.2% students nationwide, funded by the Chinese Ministry of Education
- First-Class Academic Excellence Scholarship (Three Times), December 2022, 2023, 2024
- Highest academic scholarship awarded to top 1% students in Shanghai Jiao Tong University
Selected Publications
- StraTA: Incentivizing Agentic Reinforcement Learning with Strategic Trajectory Abstraction
- Submitted to NeurIPS 2026 (First Author)
- Advisor: Prof. Wanli Ouyang & Dr. Zhenfei Yin
- Introduce an explicit trajectory-level strategy to guide long-horizon behavior during agentic RL, improving exploration and credit assignment across the entire learning process.
- CoMAS: Co-Evolving Multi-Agent Systems via Interaction Rewards
- Accepted by ICLR 2026 (First Author)
- Advisor: Prof. Wanli Ouyang & Dr. Zhenfei Yin
- Propose a novel framework that enables multiple agents to self-improve autonomously by learning from intrinsic reward signals from inter-agent interactions without external supervision.
- ComfyBench: Benchmarking LLM-based Agents in ComfyUI for Autonomously Designing Collaborative AI Systems
- Accepted by CVPR 2025 (First Author)
- Advisor: Prof. Wanli Ouyang & Dr. Lei Bai
- Build a benchmark for LLM-based agents to autonomously design collaborative AI systems in ComfyUI, along with a well-performing multi-agent framework to solve the tasks.
Internship Experience
- Shanghai Artificial Intelligence Laboratory, May 2024 - May 2026
- Trainee Researcher, AI for Science Group
- Research on the topics of LLM-based Agents, Multi-agent Systems, and Agentic RL under the supervision of Prof. Wanli Ouyang and Dr. Lei Bai, producing multiple publications at top-tier conferences.
