Junbo Li
Scholar

Junbo Li

Google Scholar ID: XVSPtCoAAAAJ
University of Texas at Austin
agentic reasoning LLMreinforcement learning
Citations & Impact
All-time
Citations
213
 
H-index
7
 
i10-index
7
 
Publications
12
 
Co-authors
0
 
Resume (English only)
Academic Achievements
  • Publications: Including ICML 2025's 'Prior-Informed Preference Alignment' (first author) and 'Hierarchical Asynchronous Local SGD'; NeurIPS 2024's 'Web2Code'; COLM 2024's 'LLM360' and 'CrystalCoder'; ICLR 2024's 'Sparse Subspace Variational Inference', etc. Awards: Recipient of the Amazon AI Ph.D. Fellowship in 2025; won a silver medal in the China Mathematical Olympiad (CMO) during high school.
Research Experience
  • Visiting student with Prof. Eric Xing from 2023 to 2024; visiting student with Prof. Ruqi Zhang in 2023; started an internship at Amazon Rufus Alignment team in May 2025.
Education
  • Ph.D.: University of Texas at Austin, Computer Science, co-advised by Prof. Atlas Wang and Prof. Qiang Liu (2024-present). M.S.: University of California, Santa Cruz, Computer Science (2023). B.S.: Fudan University, Mathematics and Applied Mathematics (2021), worked with Prof. Zhaoran Wang and Zhuoran Yang during undergraduate studies.
Background
  • Research Interests: Reasoning agentic LLM and reinforcement learning. Professional Field: Computer Science.
Miscellany
  • Personal Interests: Participated in math competitions for one year during high school.
Co-authors
0 total
Co-authors: 0 (list not available)