Towards Safe and Reliable Foundation Models

Jeonghyeon Kim is a Ph.D. student in Data Science at Seoul National University of Science and Technology (SeoulTech), where he is advised by Prof. Sangheum Hwang. His research centers on building Safe and Reliable Foundation Models. He is particularly focused on Multi-modal Representation Learning to establish robust Out-of-Distribution (OoD) detection mechanisms in open-world settings. Recently, he has expanded his research to Machine Unlearning via Mechanistic Interpretability, aiming to enable Large Language Models (LLMs) to selectively and safely forget information. His ultimate goal is to pioneer AI solutions that are not only state-of-the-art in performance but also transparent, secure, and trustworthy—empowering their adoption in mission-critical domains such as healthcare, autonomous systems, and beyond.