Haolin Liu

Hi! I am Haolin Liu, a fourth-year PhD student at the University of Virginia, where I am fortunate to be advised by Prof. Chen-Yu Wei. Prior to this, I received my bachelor’s degree in Computer Science from ShanghaiTech University, where I studied chemistry for 1.5 years before transitioning to computer science for 2.5 years.

I am interested in developing principled and practical algorithms for Reinforcement Learning (RL), and understanding their training dynamics. Recently, I mainly focus on agentic RL and RL theory.

• On the theoretical side, I study unified principles for RL algorithm design and seek to characterize the minimal structure required for sample-efficient learning. My recent works ([1], [2]) propose the most unified RL-theory frameworks to date, capable of handling both model-based and model-free RL in stationary and non-stationary environments.
• On the practical side, I aim to build self-improving LLM agents that can continually learn from interactions. Toward this goal, I am building new RL environments and algorithms for scalable long-horizon agent training.

Currently, I am an intern at Bytedance Seed in San Jose, where I work on agentic RL across environment development and algorithm design. Previously, I interned at Tencent Hunyuan (formerly Tencent AI Lab) in Seattle. I also have long-term research collaborations with Google Research.