Xun Huang (/shuun hwang/)

I am a founder of a stealth startup. Previously, I was a Research Scientist at Adobe Research, an Adjunct Professor at CMU, and a Research Scientist at NVIDIA. I obtained my PhD in Computer Science from Cornell in 2020, advised by Professor Serge Belongie.

I led interactive video world model research at Adobe and pioneered foundational technologies that is powering state-of-the-art world models, including autoregressive diffusion transformers (CausVid) and Self Forcing. At NVIDIA, I developed one of the first public text-to-image demo (GauGAN2), coauthored papers that serve as the basis of NVIDIA's text-to-image and text-to-3D foundation models and shipped them into production. During my PhD, I invented Adaptive Instance Normalization (AdaIN) and was the first to demonstrate its effectiveness in generative neural networks. AdaIN became a foundational component of StyleGAN and played a key role in the first working diffusion model. Variants of AdaIN are now used in nearly all state-of-the-art diffusion models. My PhD research was supported by Adobe Research Fellowship (2019), Snap Research Fellowship (2019), and NVIDIA Graduate Fellowship (2018).