Embedded Files
The banner image was generated via one of powerful diffusion models, SDXL, with prompt: A cozy Italian café scene with a small porcelain cup of freshly brewed espresso on a saucer, placed on a rustic wooden table. In the background, warm sunlight filters through narrow cobblestone streets lined with colorful buildings and flower boxes. A barista stands behind a vintage espresso machine, and an Italian flag flutters nearby. The atmosphere is relaxed and authentic, capturing the charm of everyday life in Italy.
Chieh-Hsin (Jesse) Lai
Chieh-Hsin (Jesse) Lai
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. He has co-organized NeurIPS EXPO workshops on media content restoration and editing (NeurIPS 2023 & 2024) and on creative and protective AI for music and entertainment (NeurIPS 2025). He has also organized and delivered tutorials on generative AI at ISMIR 2024, ICASSP 2025, and IJCNN 2025. Recently, he led and authored a book, The Principles of Diffusion Models, which provides a systematic treatment of the foundations of diffusion models.
For more information, please visit his Google Scholar and Personal Website. Please contact Chieh-Hsin (Jesse) Lai with any questions or concerns about the tutorial.
Subham Sahoo
Subham Sahoo
Subham Sahoo is a Senior Research Scientist at the MBZUAI Institute of Foundation Models, where he leads a research team focused on diffusion language models (diffusion-LLMs). He holds a Ph.D. from Cornell University, where he conducted foundational work in this area. Most notably, he proposed the MDLM framework, which has played a pivotal role in shaping the field of discrete diffusion and has been adopted at industrial scale such as ByteDance’s state-of-the-art diffusion-LLM and NVIDIA’s GenMol foundation model for drug discovery. He has delivered talks on diffusion-LLMs at venues including the ICML 2024 AccMLBio Workshop, the NeurIPS 2024 Compression Workshop, and the ICLR 2025 DeLTA Workshop.
Yuki Mitsufuji
Yuki Mitsufuji
Yuki Mitsufuji
Page updated
Google Sites
Report abuse