- Paper awards or honorable mentions at NIPS, AISTATS, KDD
- Core contributor: parallel track mixture of experts architecture (PT-MoE)
- Published multiple papers such as 'Recurrent Drafter for Fast Speculative Decoding in Large Language Models', 'Linear Complexity Randomized Self-attention Mechanism' etc.
- Open source projects with 1400+ stars on GitHub
Research Experience
- Research Scientist at Meta
- Distinguished Scientist at Apple, focusing on foundation models
Education
PhD in Computer Science from Princeton University, advised by Prof. David Blei.
Background
Research interests include foundation models, natural language processing, and machine learning. Currently a Research Scientist at Meta, previously a Distinguished Scientist at Apple.
Miscellany
Media reports: VentureBeat, SiliconANGLE, InfoQ, futurism, cnBeta, Sina Tech, iThome, ChinaEmail, eepw, QbitAI, oschina