Published high-impact papers such as 'Efficient Autoregressive Shape Generation via Octree-Based Adaptive Tokenization' (ICCV 2025)
Led development of multiple open-source projects including ControlNet, FramePack, FlashTex, Layer Diffuse, AnimateDiff, EmphasisChecker, Flipper, and the Stanford Cable TV News Analyzer
Background
Forest Baskett Professor of Computer Science at Stanford University
Director of the Brown Institute for Media Innovation at Stanford University
Consulting AI Scientist at Roblox
Research areas include computer graphics, human-computer interaction (HCI), and visualization
Focuses on how cognitive design principles can improve the effectiveness of audio/visual media, and implements these principles in interactive and automated design tools