About
My research goal is to accelerate materials discovery with AI. I am particularly interested in closed-loop discovery, where AI generates candidate materials, evaluates them through property prediction and simulation, learns from experimental feedback, and is orchestrated by AI scientists that run the loop autonomously. I am interested in every component of this loop and in connecting them into a single discovery system.
AI Scientist
AI4Mat@NeurIPS'26Generate
Propose candidate crystals
Predict & Simulate
ML potentials and molecular dynamics
Experiment
Synthesis and measurement
Learn
Correct models with measurements
↺ repeat from Generate
Publications
ExFiT-Bench: What Transfers When Universal Interatomic Potentials are Fine-Tuned on Experiments?
Kiyoung Seong, Nayoung Kim, Sungsoo Ahn
Under review at ICLR 2027
Discovering Crystal Structure Prediction Algorithms with an AI Co-Scientist
Kiyoung Seong, Nayoung Kim, Sungsoo Ahn
NeurIPS 2026 AI4Mat Workshop (Spotlight) · Under review at ICLR 2027
Multimodal Crystal Flow: Any-to-Any Modality Generation for Unified Crystal Modeling
Kiyoung Seong, Sungsoo Ahn, Sehui Han, Changyoung Park
ICML 2026
Transition Path Sampling with Improved Off-Policy Training of Diffusion Path Samplers
Kiyoung Seong, Seonghyun Park, Seonghwan Kim, Woo Youn Kim, Sungsoo Ahn
ICLR 2025
On Scalable and Efficient Training of Diffusion Samplers
Minkyu Kim*, Kiyoung Seong*, Dongyeop Woo, Sungsoo Ahn, Minsu Kim(* equal contribution)
NeurIPS 2025
Experience
- 2025.03 — Present
- 2025.09 — 2026.02
- 2023.06 — 2025.02
Researcher
POSTECH, CSE
- 2017.03 — 2023.02
B.S. in Mathematics Education (Advanced Mathematics Track)
Korea University