Published Spurious Rewards: Rethinking Training Signals in RLVR (2025-05)
Released ReasonIR, a retriever specifically trained for reasoning tasks (2025-04)
Launched OpenScholar, an AI assistant proficient at answering research questions with accurate paper citations using online paper resources (2024-11)
Introduced MassiveDS-1.4T, the first open-source trillion-token datastore and released codes for distributed indexing (2024-07)
Proposed DistFlashAtten, a distributed memory-efficient attention with sequence parallelism for Long-context LLMs training (2023-10)
Research Experience
Visiting researcher at Meta, working with Scott Yih and Mike Lewis.
Education
Currently a second-year PhD student at the University of Washington, advised by Prof. Pang Wei Koh and Prof. Luke Zettlemoyer. Completed a master's degree in Machine Learning at CMU, advised by Prof. Eric Xing, and an undergraduate degree in Mathematics at XJTU.
Background
Interested in building an AI system that can automatically collect information from diverse sources, conduct deliberate practice, and decide how to spend time, compute, and storage to handle complex tasks.
Miscellany
Personal links: Google Scholar / GitHub / Twitter / LinkedIn / Email