Hamin Koo

hamin_temp__.png

Hello! I am a PhD student in AI at Yonsei University, advised by Prof. Jaehyung Kim in the Machine and Language Learning Lab (ML3).

I’m interested in understanding and improving the reliability and alignment of large language models (LLMs) in real-world interactions. I study systematic and unintended model behaviors—including biases, safety failures, and behaviors that shape human judgment and decision-making—and develop methods to evaluate and mitigate them.

News

Oct 03, 2026 HUB-J is accepted to NeurIPS 2026! Paper coming soon 🎉
Sep 04, 2026 I’ll be joining the UCI NLP Group as a visiting researcher, working with Prof. Sameer Singh! 🇺🇸
Apr 07, 2026 EMCEE is accepted to ACL 2026 as an oral presentation! See you in San Diego 🇺🇸

Publications

* denotes equal contribution

2026

  1. LLM-as-a-Judge, Bias
    hubj_figure.png
    Auditing the Judge: Human-Grounded Bias Discovery, Quantification, and Mitigation in LLM Judges
    Hamin Koo*, Chanjoo Jung*, Fangzhao Wu, and Jaehyung Kim
    Advances in Neural Information Processing Systems (NeurIPS), 2026
  2. Multilingual
    emcee_figure.png
    Extracting and Emulsifying Cultural Explanation to Improve Multilingual Capability of LLMs
    Hamin Koo and Jaehyung Kim
    Annual Meeting of the Association for Computational Linguistics (ACL), Oral Presentation, 2026
    PDF Code
  3. Jailbreak, Optimization
    amis_figure.png
    Align to Misalign: Automatic LLM Jailbreak with Meta-Optimized LLM Judges
    Hamin Koo, Minseon Kim, and Jaehyung Kim
    International Conference on Learning Representations (ICLR), 2026
    PDF Code Website

2024

  1. RAG, Optimization
    qoqa_figure.png
    Optimizing Query Generation for Enhanced Document Retrieval in RAG
    Hamin Koo, Minseon Kim, and Sungju Hwang
    Preprint, 2024
    PDF