MATHEMATICS → GENERATIVE INTELLIGENCE
Ph.D. Student in Artificial Intelligence
Yonsei University · Visual Intelligence Lab
Learning to generate.
Learning to understand.
Diffusion, flows, and predictive states.
From mathematical structure to plausible worlds.
Seoul, Republic of Korea
RESEARCH PORTFOLIO / 2026A mathematical perspective on generative intelligence.
I build generative models that do more than create compelling images. I want them to capture structure, identity, and how the world evolves.
I'm a Ph.D. student in Artificial Intelligence at Yonsei University, advised by Prof. Youngjung Uh in the Visual Intelligence Lab. My background is in mathematics, from harmonic analysis to neural tangent kernels.
My research spans diffusion and flow-based generation, semantic control, and consistent generation. Most recently, at Microsoft Research Asia, I explored internal predictive states for physically plausible video world models.
Geometry, controllability, and worlds that make sense.
2025 — 2026Showing all 5 research works.
An internal predictive state that evolves alongside video representations, bringing physically plausible futures into focus before denoising is complete.
Balancing real-image inversion trajectories and generated pairs to reduce synthetic-data bias in rectified-flow training.
A geometric view of classifier-free guidance: damp misaligned tangential components to keep generation closer to its intended condition.
Training-free semantic feature control that balances a consistent identity with the freedom of each individual scene.
Research on trigger conditioning, capability leakage, and isolation in personalized diffusion models.
* Equal contribution. Published papers and submissions are labeled separately. Figures are from the respective project pages; click to view full size.
Research Intern
Multimodal Reasoning Group · Beijing, China
Worked with Jianlong Fu on video world models and future-predictive representations. Developed an internal predictive state for video diffusion transformers to improve the physical plausibility of generated futures and enable early candidate selection.
Research Intern
Yonsei University · Seoul, Republic of Korea
Yonsei University
Visual Intelligence Lab · Advisor: Prof. Youngjung Uh
Diffusion and flow-based generative models, identity-consistent generation, and predictive states for video world models.
Yonsei University
Advisor: Prof. Sungchul Lee
Thesis: Neural Tangent Matrix Factorization Recommendation System. Studied harmonic analysis and oscillatory integrals; explored neural tangent kernels and graph-based machine learning.
Chonnam National University
Gwangju, Republic of Korea
CVPR 2026
Reviewer for leading conferences in computer vision and machine learning.
