About me 👋
I am a second-year Computer Science PhD student at UC Berkeley, advised by Matei Zaharia and Sewon Min. I'm a member of Sky Computing Lab, Berkeley AI Research (BAIR) and Berkeley NLP Group.
My current research focuses on understanding and improving the reasoning capabilities of large language models, especially on open-ended, long-tail, research-level problems. I am particularly interested in novel learning methods that are more data-efficient and enable models to reason longer and more intelligently, as well as evaluation methods that can reliably assess their reasoning quality.
Before Berkeley, I earned my BS in Computer Science from Nanjing University. I worked on distributed systems (reliability and verification) and program analysis and have won an OSDI Best Paper Award. I still enjoy talking about those topics—just don't blame me if I mislead you.
If you are interested in my research, feel free to reach out by email. If you are an undergraduate student who would like to work together, I recommend taking at least one foundational ML course first (such as CS 288, CS 285, or CS 289).
Selected Publications
- Reliable Fine-Grained Evaluation of Natural Language Math Proofs
Wenjie Ma, Andrei Cojocaru, Neel Kolhe, Bradley Louie, Robin Said Sharif, Haihan Zhang, Vincent Zhuang, Matei Zaharia, Sewon Min
[Paper] [Website] [Dataset] Preprint, 2025 - Reasoning Models Can Be Effective Without Thinking
Wenjie Ma, Jingxuan He, Charlie Snell, Tyler Griggs, Sewon Min, Matei Zaharia
[Paper]
Preprint, 2025 - Anvil: Verifying Liveness of Cluster Management Controllers
Xudong Sun, Wenjie Ma, Jiawei Tyler Gu, Zicheng Ma, Tej Chajed, Jon Howell, Andrea Lattuada, Oded Padon, Lalith Suresh, Adriana Szekeres, Tianyin Xu
In Proceedings of the 18th USENIX Symposium on Operating Systems Design and Implementation (OSDI 2024), Santa Clara, CA, Jul. 2024
[Paper] [Code] [Talk]
🏆 Jay Lepreau Best Paper Award - Context Sensitivity without Contexts: A Cut-Shortcut Approach to Fast and Precise Pointer Analysis
Wenjie Ma*, Shengyuan Yang*, Tian Tan, Xiaoxing Ma, Chang Xu, Yue Li
In Proceedings of the 44th ACM SIGPLAN Conference on Programming Language Design and Implementation (PLDI 2023), Orlando, FL, June 2023
[Paper] [Code]
Miscellaneous
My hobbies tend to evolve over time. Lately, I've been really into everything related to food and drinks—experimenting with recipes and ingredients, and occasionally trying to reverse-engineer restaurant dishes at home.
I enjoy conversations with people from diverse fields, especially those I'm less familiar with. My research is problem-driven, and I like interdisciplinary work.
I love learning about different cultures and seize every chance I get to travel.