Generative Intelligence Lab

Welcome to the Generative Intelligence Lab (GIL)!

 Our interest is to push the boundaries of modern generative AI, pioneering technologies that serve as key solutions to real-world challenges.

We dedicate to accelerate the speed of generative AI systems with physical consistency and theoretical formality, eventually aiming to create

a physically trustful, rapid physical-twin system that can simulate, solve, adjust to challenging real-world dynamics.

News

• [2026.03] Our lab has received the Early-career Researcher Program [신진연구 A] grant for a project on physics-consistent AI dynamical systems!

• [2026.02] Our lab member Jongwon Lee have been selected as recipients of the Insan Scholarship from the Insan Scholarship & Cultural Foundation.

• [2026.01] A paper 'Mean-Field Neural Differential Equations: A Game-Theoretic Approach to Sequence Prediction' has been accepted to ICLR 2026.

• [2025.09] A paper 'Neural Hamiltonian Diffusions for Modeling Structured Geometric Dynamics' has been accepted to NeurIPS 2025.

• [2025.09] A paper 'Poisson-Algebraic Parallel Scan: A Fast Symplectic Framework for Neural Hamiltonians' has been accepted to Neurreps workshop in NeurIPS 2025.

• [2025.02] Our Generative Intelligence Lab is now open! 🎉🎉