- ImpRAG: Retrieval-Augmented Generation with Implicit Queries
- Seq2seq is All You Need for Coreference Resolution
- Improving Multitask Retrieval by Promoting Task Specialization
- EntQA: Entity Linking as Question Answering
- Understanding Hard Negatives in Noise Contrastive Estimation
Research Experience
Research Intern at Microsoft, Jun 2022 – Sep 2022, worked with Prof. Chenyan Xiong and Arnold Overwijk on multi-task retrieval; Research Intern at Meta FAIR, May 2024 – Dec 2024, worked with Dr. Mingda Chen, Dr. Scott Yih, and Dr. Victoria Lin on query-free RAG.
Education
Currently a PhD candidate in Computer Science at Rutgers University, advised by Prof. Karl Stratos; received an M.S. in Computer Science from Rutgers in Spring 2021; earned a B.S. in Applied Physics from the University of Science and Technology of China in 2019.
Background
Research interests include retrieval-augmented generation and entity-centric language understanding. Particularly interested in the reasoning capabilities of large language models (LLMs).
Miscellany
Personal website powered by Jekyll & AcademicPages, a fork of Minimal Mistakes.