Hyomin Lee

I'm a master's student in the MLAI Lab at KAIST, under the supervision of Prof. Sung Ju Hwang. I earned my bachelor's degree in Cyber Defense and Artificial Intelligence from Korea University.

My primary research interests lie at the intersection of security and artificial intelligence, with a particular focus on:

  • Model Robustness
  • Safe & Trustworthy AI
  • LLM Agent Security

Education

  • M.S. in Artificial Intelligence, KAIST
    • Sep 2025 - Present
  • B.E. in Cyber Defense & Artificial Intelligence, Korea University
    • Mar 2021 - Feb 2025

Publications

  • T-MAP: Red-Teaming LLM Agents with Trajectory-aware Evolutionary Search
    • Hyomin Lee, Sangwoo Park, Yumin Choi, Sohyun An, Seanie Lee, Sung Ju Hwang
    • ICLR Workshop on Agents in the Wild: Safety, Security, and Beyond, 2026
  • Enhancing Variational Autoencoders with Smooth Robust Latent Encoding
    • Hyomin Lee*, Minseon Kim*, Sangwon Jang, Jongheon Jeong, Sung Ju Hwang
    • arXiv, 2025
  • Automatic Jailbreaking of the Text-to-Image Generative AI Systems
    • Minseon Kim, Hyomin Lee, Boqing Gong, Huishuai Zhang, Sung Ju Hwang
    • ICML Worshop on NextGenAISafety, 2024

Honors and Awards

  • 1st place, AI Cyber Defense Contest (ACDC) University Division, 2025
  • Presidential Science Scholarship, 2023-2024
  • Best Prize, LG U+ Seucurity Hackathon, 2024
  • Best Prize, Hacktheon University Student Cybersecurity Competition, 2022
  • Top 10 (Certified by the Minister of Science and ICT), KITRI Best of the Best 9th, 2021

Contact