PhD student in the Department of Computing at The Hong Kong Polytechnic University.

My research focuses on reinforcement learning with human feedback, reward modeling, reasoning in large language models, and embodied AI. I am interested in building learning-based agents that are more reliable, adaptive, and practically useful.

I use this GitHub profile as a concise index to my research code, publications, and academic homepage.

For a fuller academic profile, including publications, scholarly metrics, writing, and contact information, please visit my personal website:

• Academic homepage

• Reinforcement learning with human feedback
• Reward modeling and feedback-driven learning
• Reasoning in large language models
• Embodied AI and learning-based agents

• Google Scholar
• Academic homepage
• Email

• C3 Contextual Counterfactual Credit Assignment for multi-agent reinforcement learning in LLM collaboration.

• AccuracyParadox-RLHF Research code for studying when stronger reward models do not necessarily yield better RLHF outcomes.

• battam1111.github.io Source code for my academic homepage and writing.

• Email: yan-jun.chen@connect.polyu.hk
• Location: Hong Kong