Minseong Bae

I am a M.S. student at KAIST MLV Lab (School of Computing), advised by Prof. Hyunwoo J. Kim. Previously, I received my B.E. in Computer Science and Engineering and Mathematics from Korea University in 2025. For more details, please see my CV.

My research interests lie in machine learning, with a particular focus on geometric deep learning, diffusion/flow-based generative models, and multimodal learning. Currently, I’m focusing on RL-based post-training methods for generative models.

Beyond theory, I am deeply interested in applying machine learning to natural sciences to tackle impactful real-world problems (AI4Science). In this context, I have been working on multimodal language models, diffusion-based generative models and agentic systems for biomolecules and weather patterns.

My long-term research goal is to develop robust, scalable learning methods that enable self-evolving, human-collaborative agents for scientific discovery.

Feel free to reach out via email if you’d like to discuss anything related to me!