Title： What Do Diffusion Models Actually Learn?
Date：2026/10/23 14:20-15:30
Location：R103, CSIE
Speaker：Dr. Ya-Ping Hsieh, Academia Sinica
Host：Prof. Yen-Huan Li
Abstract:
Diffusion models are usually presented as tools for learning complex data distributions. In this talk, I will discuss a complementary view: before learning the full distribution, diffusion models first learn where the data lives. In continuous spaces, this means recovering the geometry of the data manifold before accurately modeling the density on it. In discrete spaces, such as language, it means learning validity or support structure before learning fine-grained frequencies. This perspective helps explain why diffusion models can generate realistic and novel samples even when their learned scores are still coarse. The talk will introduce the main intuition behind this “geometry-before-density” principle, explain how it appears in both continuous and discrete diffusion models, and discuss its implications for generalization, memorization, and the design of more robust generative algorithms.
Bio:
Ya-Ping Hsieh is an Assistant Research Fellow at the Institute of Statistical Science, Academia Sinica. His research focuses on the mathematical foundations of machine learning, particularly optimization and generative models.