Zongxia Li
Scholar

Zongxia Li

Google Scholar ID: X3uWrikAAAAJ
University of Maryland, College Park
Natural Language ProcessingMultimodal Models
Citations & Impact
All-time
Citations
808
 
H-index
8
 
i10-index
8
 
Publications
16
 
Co-authors
6
list available
Resume (English only)
Academic Achievements
  • - Awards: Lambda Lab Research Compute Grant (2025), NIST Award Fellowship (2023-2025)
  • - Publications: 'Self-Rewarding Vision-Language Model via Reasoning Decomposition' (Preprint, 2025), proposing a new method for training VLMs, i.e., 'look first, then reason'
Research Experience
  • - 2025: Interned at Tencent AI Lab, mentored by Wenhao Yu, collaborated with Chengsong Huang, worked on Self-evolving VLMs and LLMs
  • - 2024: Interned at Adobe Document Intelligence Lab, focusing on improving LLM automatic evaluations to improve downstream training
Education
  • - Ph.D. Student: University of Maryland, College Park, Department of Computer Science, Sep. 2022 - present, Advisor: Jordan Boyd-Graber
  • - B.S. in Computer Science and Mathematics (Honors): University of Maryland, College Park, Sep. 2018 - May 2022
Background
  • - Research Interests: Human-Centered AI, Multimodal Models, Post-Training and Evaluation
  • - Field: Computer Science
  • - Brief Introduction: A fourth-year Ph.D. candidate at the Department of Computer Science, University of Maryland, College Park, advised by Jordan Boyd-Graber.
Miscellany
  • Reviewer for ACL, EMNLP, NAACL, ICLR, NeurIPS, CVPR, IEEE, ACM