Papers & manuscripts

Papers below are currently under review. Author order follows contribution; my name is bolded.

1. 03
   [Paper 3]

   Jimin Yeom, Jonghyun Hong, Sungyoon Lee

   Under review·2026·AI Safety / LLM Jailbreaking / Optimization / Continuous Defense
2. 02
   [Certifying Robustness Large Language Models via Discrete-Continuous Randomized Smoothing]

   Subin Jang, Jimin Yeom, Sungyoon Lee

   Trustworthy AI for Good (AI4GOOD) @ICML 2026·2026·AI Safety / LLM Jailbreaking / Certified Defense
3. 01
   [A Training-Dynamics View of Catastrophic Overfitting: Understanding and Prevention]

   Jimin Yeom, Sungyoon Lee

   Trustworthy AI for Good (AI4GOOD) @ICML 2026·2026·AI Safety / Adversarial Training / Catastrophic Overfitting / Training Dynamics