About Me

I am an Integrated M.S.–Ph.D. student at Hanyang University (HYU NLP Lab), advised by Prof. Taeuk Kim.

My research focus has recently shifted toward LLM-based Agents. I am interested in how agents use tools over long interactions, and in what they actually retain — and why — as they do so.

  • LLM Agents: Multi-turn tool use, and agent memory studied at the mechanistic level — why different designs behave differently, not just how they score.
  • Reliable NLP: Knowledge editing, unlearning, interpretability, and multilingual safety (Previous focus).

Selected Publications

Query Lens: Interpreting Sparse Key-Value Features with Indirect Effects
Hwiyeong Lee, Ingyu Bang, Uiji Hwang, Hyelim Lim, Taeuk Kim
ICML 2026 NEW
Development and Evaluation of a Dual-Expertise, Utterance-Level Framework for LLM-Based Science Classroom Discourse Analysis
Jin Eun Yoo, Nam-Hwa Kang, Suna Ryu, Jun-ki Lee, Youngsun Kwak, Taeuk Kim, Hyeong Gwan Kim, Youngwoo Shin, Uiji Hwang
LAK 2026
A Survey on Machine Unlearning for Large Language Models
Uiji Hwang, Taeuk Kim
Communications of the KIISE, Vol. 43, No. 11.
Domestic Journal
Memorization or Reasoning? Exploring the Idiom Understanding of LLMs
Jisu Kim*, Youngwoo Shin*, Uiji Hwang, Jihun Choi, Richeng Xuan, Taeuk Kim
EMNLP 2025

Teaching Experience

  • Natural Language Processing (Teaching Assistant)
    Spring 2026
  • Artificial Intelligence (Teaching Assistant)
    Fall 2025

Projects

Teaching Feedback Analytics (KNUE) 2025
Developed a dual-expertise framework for analyzing science classroom discourse using LLMs. (Project Completed)

News

May 2026 Paper accepted to ICML 2026.
Jan 2026 Paper accepted to LAK 2026.
Nov 2025 Survey on Machine Unlearning published in KIISE.
Aug 2025 Two papers accepted to EMNLP 2025 Main.

Education

Hanyang University Mar 2025 – Present
Integrated M.S.–Ph.D. in Computer Science (HYU NLP Lab)
Chung-Ang University Mar 2018 – Feb 2025
B.A. in German Language & Literature / B.S. in AI German Processing