Yilun Xu

Reserach Scientist at Meta

Cambridge, Massachusetts, United States
email-iconphone-icongithub-logolinkedin-logotwitter-logostackoverflow-logofacebook-logo
Join Prog.AI to see contacts
email-iconphone-icongithub-logolinkedin-logotwitter-logostackoverflow-logofacebook-logo
Join Prog.AI to see contacts

Summary

👤
Senior
🎓
Top School
Yilun Xu is a research scientist with eight years of experience advancing generative modeling and diffusion techniques across industry and academia. He completed a PhD-level trajectory at MIT and moved from influential academic work at Stanford (ICLR oral and follow-ups) into research roles at NVIDIA, Google DeepMind (Gemini team), and now Meta. At NVIDIA he co-led diffusion distillation efforts such as COSMOS NANO, and at DeepMind worked on generative models that simulate the world. His work bridges theoretical contributions (V-information, anytime sampling) with practical model distillation and large-scale generative systems. Based in Cambridge, MA, he combines deep probabilistic modeling expertise with hands-on engineering to push sampled-based methods toward production. Colleagues know him for pairing rigorous research results with reproducible code and pragmatic distillation strategies that speed up deployment.
code8 years of coding experience
job1 year of employment as a software developer
bookDoctor of Philosophy - PhD, EECS, Doctor of Philosophy - PhD, EECS at Massachusetts Institute of Technology
bookBachelor's degree, Computer Science, Bachelor's degree, Computer Science at Peking University
github-logo-circle

Github Skills (42)

generative-adversarial-network10
generative-model10
text-to-image9
multi-task-learning9
loss9
dmi8
diffusion-models8
survey8
pytorch8
taxonomy7
classifier7
flux7
deep-learning7
noise7
autoregressive-models6

Programming languages (2)

Jupyter NotebookPython

Github contributions (5)

github-logo-circle
Newbeeer/pfgmpp

Dec 2022 - Mar 2023

Code for ICML 2023 paper, "PFGM++: Unlocking the Potential of Physics-Inspired Generative Models"
Contributions:120 commits, 1 PR, 41 pushes in 2 months
generative-model
Newbeeer/Poisson_flow

Oct 2022 - Jan 2023

Code for NeurIPS 2022 Paper, "Poisson Flow Generative Models" (PFGM)
Contributions:68 commits, 108 pushes, 1 branch in 2 months
generative-model
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.
Request Free Trial