Jui-Hsien Wang

Research Scientist/Engineer
Adobe Research, Seattle



Email: juiwang at adobe dot com
Office: 801 N 34th St, Seattle, WA
About

I received my Ph.D. in computational mathematics from Stanford University in 2020, where I was advised by Doug James. I also have a MS degree from UCSD/Cornell and a BS from National Taiwan University.

My research focuses on modeling the world. I study generative models and numerical simulations to help build multimodal representations of the world with deep mathematical insights. Here are some recent topics I am interested in:

  • Diffusion models (image & video)
  • Video understanding
  • Multimodal machine perception and reasoning
  • Physics-based modeling, simulation, and optimization
  • Physics-based sound synthesis

-- Study hard what interests you the most in the most undisciplined, irreverent and original manner possible. -- Richard Feynman

Collaboration: I am currently taking students for the 2025 summer internship. If you are interested, please reach out.

My Chinese name is 王睿賢. I go by "Jui" which sounds like "Ray" in English. If you'd like to try to pronounce the syllables after the hyphen, try it with a silent "H". Here is how I pronounce my first name when talking in English:

News

  • Oct 15 2024: Firefly Video (T2V/I2V) and GenExtend, two features I worked on for multiple years, are announced!
  • Sep 19 2024: 1 SIGGRAPH Asia paper accepted. Congrats Kangrui!
  • Apr 26 2024: GenLens won the best paper award at PacificVis. Congrats Tica!
  • Feb 27 2024: 1 paper accepted to PacificVis (GenLens: sysmetric evaluation of visual GenAI model outputs).
  • Feb 27 2024: VideoMosaic paper is accepted to CVPR 2024.
Industrial Highlights

Publications

More Projects

Under construction ...