Kiyoung Seong

M.Sc.student, KAIST

kyseong98@kaist.ac.kr

Bio

My research goal is to develop AI tools for drug discovery and material design. To be specific, I am focusing on (1) accelerating molecular dynamics using generative models [ICLR'24] [ICLR'25] and ML collective variables [ICML'25], and (2) advancing generative model [Arxiv'25] [Arxiv'25] for scientific discovery. I'm a graduate student at KAIST AI, advised by Sungsoo Ahn.

Publications

Most recent publications on Google Scholar.

Transition Path Sampling with Improved Off-Policy Training of Diffusion Path Samplers

Kiyoung Seong, Seonghyun Park, Seonghwan Kim, Woo Youn Kim, Sungsoo Ahn

ICLR'25: International Conference on Learning Representations, 2025.

Paper Code Project Page

Learning Collective Variables from Time-lagged Generation

Seonghyun Park, Kiyoung Seong, Soojung Yang, Rafael Gomez-Bombarelli, Sungsoo Ahn

ICML'25: International Conference on Machine Learning (GenBio workshop), 2025.

Paper

On scalable and efficient training of diffusion samplers

Minkyu Kim, Kiyoung Seong, Dongyeop Woo, Sungsoo Ahn, Minsu Kim

Arxiv'25: arXiv preprint, 2025.

Paper

Energy-based generator matching: A neural sampler for general state space

Dongyeop Woo, Minsu Kim, Minkyu Kim, Kiyoung Seong, Sungsoo Ahn

Arxiv'25: arXiv preprint, 2025.

Paper

Collective Variable Free Transition Path Sampling with Generative Flow Network

Kiyoung Seong, Seonghyun Park, Seonghwan Kim, Woo Youn Kim, Sungsoo Ahn

ICML'24: International Conference on Machine Learning (SPIGM workshop), 2024.

Paper

Vitæ

Kiyoung Seong
kyseong98@kaist.ac.kr