Yang Song

I am the Research Principal at Meta Superintelligence Labs (MSL), where I work with Chief Scientist Shengjia Zhao to help shape the lab’s research direction.

During my Ph.D., I developed some of the core concepts and techniques behind score-based diffusion models, which now power many generative AI systems for images, video, and audio. You can read more about this line of work in a blog post or a Quanta Magazine article. My research has been recognized with an ICLR Outstanding Paper Award and Ph.D. fellowships from Apple and J.P. Morgan.

Before joining Meta, I led the strategic explorations team at OpenAI. I received my Ph.D. in Computer Science from Stanford University, advised by Stefano Ermon.