Scholar
Zhoujun Cheng
Google Scholar ID: t41vrrQAAAAJ
UC San Diego
Citations & Impact
Citations
1,759
H-index
13
i10-index
17
Publications
20
Co-authors
13
list available
Publications
Cited
0
Resume (English only)
Academic Achievements
- Revisiting Reinforcement Learning for LLM Reasoning from A Cross-Domain Perspective
- MegaMath: Pushing the Limits of Open Math Corpora
- OSWorld: Benchmarking Multimodal Agents for Open-Ended Tasks in Real Computer Environments
- What Are Tools Anyway? A Survey from the Language Model Perspective
- OpenAgents: An Open Platform for Language Agents in the Wild
- Lemur: Harmonizing Natural Language and Code for Language Agents
- Binding Language Models in Symbolic Languages
- HiTab: A Hierarchical Table Dataset for Question Answering and Natural Language Generation
- FORTAP: Using Formulae for Numerical-Reasoning-Aware Table Pretraining
Research Experience
- Conducted research at HKUNLP and Microsoft Research Asia.
Education
- Currently a Ph.D. student at UC San Diego, advised by Prof. Zhiting Hu; previously received B.E. and M.E. degrees in Computer Science (IEEE class) from Shanghai Jiao Tong University, and had research experiences at HKUNLP advised by Prof. Tao Yu and Microsoft Research Asia with Haoyu Dong.
Background
- Research interests include natural language processing, particularly large language model (LLM) reasoning and their interaction with the digital world. Before the LLM era, focused on structured data reasoning over tables and spreadsheets.
