Weiyao Wang

Bio

I am a AI Reseasrcher/ Tech Lead at Meta Superintelligence Lab (MSL).  My research interests focus on developing multimodal foundation models, 3D world modeling systems, and large-scale perception architectures bridging research and production AI. Prior to MSL, I was at FAIR Meta, where I am fortunate to collaborate and receive mentorship from Dr. Du Tran. Prior to Facebook (aka Meta), I received my Bachelor degree in mathematics and computer science from Duke University, with great advisory by Prof. Rong Ge and Prof. Robert Calderbank. 

News

• 11/19/2025: We launched SAM3D! Great to be part of the project [blog][github].

• 02/26/2024: ICON accepted at CVPR 2024! Take a look here [link].

• 09/22/2023: Two papers accepted at NeurIPS 2023!

• 07/13/2023: SiLK accepted at ICCV 2023!

• 04/14/2023: Released SiLK codebase for simple learned keypoints [link].

• 02/23/2023: EgoTracks released as part of Ego4D [link].

• 04/11/2022: Generic Grouping paper accepted by CVPR 2022!

• 10/16/2021: Co-organized the Open-World Video Object Detection and Segmentation Challenge at ICCV'21 [link].