👋 About Me

I am a Master’s student in Artificial Intelligence at Hanyang University and a member of HYU NLP Lab, advised by Prof. Taeuk Kim. My research is at the intersection of Natural Language Processing (NLP), Human-Computer Interaction (HCI), and empirical AI alignment. I am interested in understanding when LLM-based systems can reliably reason, remember, and make decisions in ways that reflect human preferences. My recent research focuses on LLM-as-a-Judge, rubric-based evaluation, reward signal reliability, and memory-based preference inference for tool-calling agents. In particular, I study whether evaluation and reward signals faithfully capture human judgment, or whether they simply reproduce the prior preferences and biases of the evaluator model. My broader goal is to develop human-centered language technologies that are not only capable and interactive, but also more reliable, interpretable, and aligned with the people who use them.

🧠 Research Interests

• LLM evaluation and LLM-as-a-Judge

• Rubric-based evaluation and reward signal reliability

• Human-aligned evaluation for reasoning, dialogue, and debate

• Memory-augmented LLM agents and multi-session preference inference

• Multi-agent debate and collaborative reasoning systems

• Tool-using agents and personalized decision-making

• Human-centered dialogue systems across language and speech modalities