winter_pro.png

Minsung Kim

Ph.D. Student · Machine Intelligence Lab, SNU

Hi! I am a Ph.D. student at the Machine Intelligence Lab at Seoul National University, advised by Prof. Kyomin Jung.

I study how language models can best use what they know and external information. I have also studied how models acquire knowledge and how it can be edited or erased.

News

Aug 24, 2026 Our PRAGMA paper is accepted to EMNLP 2026!
Apr 07, 2026 Two papers accepted to ACL 2026: LLM training data and multimodal unlearning (Findings).
Mar 03, 2026 Started serving as 🇰🇷 Technical Research Personnel for mandatory military service.
Jan 22, 2026 Two papers accepted to ICLR 2026: Bilinear Relational Structure and Erase or Hide!
Sep 20, 2025 Our FaithUn paper is accepted to EMNLP 2025!

Education

M.S/Ph.D · Electrical and Computer Engineering
Seoul National University, Seoul, Korea
2023 Mar - Present
B.S · Electrical and Computer Engineering
Seoul National University, Seoul, Korea
2018 Mar - 2023 Feb

Experience

Research Intern · Host: Meeyoung Cha
Max Planck Institute for Security and Privacy
2024 Oct - 2024 Dec

Selected Publications

  1. ACL 2026
    How Training Data Shapes the Use of Parametric and In-Context Knowledge in Language Models
    Minsung Kim, Dong-Kyum Kim, Jea Kwon, Nakyeong Yang, Kyomin Jung, and Meeyoung Cha
  2. ACL 2026 Findings
    Rethinking Post-Unlearning Behavior of Large Vision-Language Models
    Minsung Kim, Nakyeong Yang, and Kyomin Jung

For more of my work, see the full publication list or my Google Scholar profile.