Zhepei “Bruce” Wei (魏哲培)
 |
Ph.D. Student Email: zhepei.wei [AT] virginia [DOT] edu |
About me
Hello! My name is Zhepei (/dʒɜ:peɪ/) Wei, and I also go by Bruce. I'm a CS PhD student at the University of Virginia, advised by Prof. Yu Meng. Prior to this, I received my B.S. and M.S. degrees in Computer Science from Jilin University in 2019 and 2022. During the research journey, I was fortunate to have been advised by Prof. Hongning Wang and Prof. Yi Chang. I'm also a former research intern at Huawei Research.
For more information, please check my CV.
What's New
[09/2024] Thrilled to receive the Copenhaver Charitable Trust Bicentennial Fellowship!
[06/2024] Check out our latest work on explicit denoising for RAG “InstructRAG: Instructing Retrieval-Augmented Generation via Self-Synthesized Rationales”!
[05/2024] Thrilled to receive the John A. Stankovic Outstanding Graduate Research Award!
[01/2024] Our paper "Incentivized Truthful Communication for Federated Bandits" is accepted at ICLR 2024.
[09/2023] Our paper "Incentivized Communication for Federated Bandits" is accepted at NeurIPS 2023.
[10/2022] Our paper "Learning Semantic Textual Similarity via Topic-informed Discrete Latent Variables" is accepted at EMNLP 2022.
[09/2022] Our paper "Towards Unified Representations of Knowledge Graph and Expert Rules for Machine Learning and Reasoning" is accepted at AACL-IJCNLP 2022.
[08/2022] Moving to Charlottesville, and starting my PhD journey at UVa CS!
[04/2022] Our paper "AttExplainer: Explain Transformer via Attention by Reinforcement Learning" is accepted at IJCAI-ECAI 2022.
Research Interests
My current research interests focus on the learning foundation of large language models (LLMs) and their applications in solving practical problems (e.g., knowledge reasoning, question answering). My recent work contributes to efficient and trustworthy language generation and reasoning with LLMs, including:
LLM alignment (e.g, self-alignment optimization)
Efficient LLM inference (e.g., ParaDecode)
Retrieval-augmented generation (e.g., InstructRAG)
I also introduced the incentivization problem in multi-agent decision-making (i.e., federated bandits) and proposed efficient communication algorithms with rigorous mathematical guarantees (NeurIPS 2023, ICLR 2024).
Selected Awards & Honors
2024. Copenhaver Charitable Trust Bicentennial Fellowship, University of Virginia.
2024. John A. Stankovic Graduate Research Award, University of Virginia.
2023. NeurIPS Scholar Award, NeurIPS Foundation.
2022. Computer Science Scholar Fellowship, University of Virginia.
2020. National Scholarship, Ministry of Education of People's Republic of China.
2020. First-Class Scholarship, Jilin University.
2020. Outstanding Graduate Award, Jilin University.
2019. Honor Graduation Award, Jilin University.
2018. National Scholarship, Ministry of Education of People's Republic of China.
2017. First-Class Scholarship, Jilin University.
2016. National Endeavor Scholarship, Ministry of Education of People's Republic of China.
2016~2017. Undergraduate Leadership Award, Jilin University.
2016~2018. Outstanding Undergraduate Award, Jilin University.