Xi Yin (殷希)

I am a Member of Technical Staff at Physical Intelligence where I work on training foundation models for Robotics. Befor that, I was a Senior Staff Research Scientist at Meta Super Intelligence Labs (MSL). I spent 6 years at Meta working on video generation as core contributors to Native Video Editing, MovieGen, Emu-Video, Make-A-Video. Prior joining Meta, I was a Senior Applied Scientist at Microsoft Cloud and AI from 2018 to 2020. I received my Ph.D. degree from the Department of Computer Science and Engineering at Michigan State University in 2018, where I was a member of the Computer Vision Lab lead by Professor Xiaoming Liu. In 2013, I graduated with a Bachelor’s degree in Electrical Engineering from Wuhan University.
I love training models and curating data. Please check my Google Scholar for a list of publications.

News

2026: I joined Physical Intelligence as a Member of Technical Staff.
2026: Invited talk at UNC: Video Generation Through the Data Lens.
2026: Core contributor to our native video editing model that launched to Meta AI and Edits where I lead the video editing data, caption model, and prompt enhancement efforts.
2024: Core contributor for Movie Gen where I was mostly involved in text-to-video pretrain, pretrain data curation, and super-resolution workstreams.
2023: Check out our recent video generation work: Emu Video.
2022: Core contributor for Make-A-Video.
2021: Featured in ICCV Daily News and Best of ICCV.
2020: Joined Facebook AI as a Research Scientist.
2018: Defensed my Ph.D. dissertation and will join Microsoft as an Applied Scientist.
2018: Joined Tencent Youtu Lab in Shanghai as a visiting researcher.
2018: Won the First Place of Fitch Beach Outstanding Graduate Research Award Competition (one winner in entire college).
2018: Won the 2018 Outstanding CSE Graduate Student Award (one winner in entire department).