The Principles of Diffusion Models: Towards Fast and Scalable Diffusion LLMs

Chieh-Hsin (Jesse) Lai received his Ph.D. in Mathematics from the University of Minnesota in 2021. He is currently a Staff Research Scientist at Sony AI and a Visiting Assistant Professor in the Department of Applied Mathematics at National Yang Ming Chiao Tung University (Taiwan). His research focuses on deep generative modeling (with an emphasis on diffusion models) and AI for science. Recently, he led and authored a book, The Principles of Diffusion Models, which provides a systematic treatment of the foundations of diffusion models.

Subham Sahoo is a Senior Research Scientist at the MBZUAI Institute of Foundation Models (IFM), where he leads the Diffusion LLM team. His research laid key foundations for the recent success of diffusion-LLMs. He received his Ph.D. from Cornell Tech with the thesis Foundations of Diffusion Language Models.

Stefano Ermon is an Associate Professor of Computer Science at Stanford University, where he is affiliated with the Stanford Artificial Intelligence Laboratory. His research centers on probabilistic modeling and generative AI, and his group has made foundational contributions to score-based and diffusion models, including score matching with Langevin dynamics, Score SDEs, and DDIM. He is also a co-founder of Inception, which builds diffusion-based large language models.