Scholar
Lei Li
Google Scholar ID: MeV4GGsAAAAJ
University of Hong Kong, Peking University
Citations & Impact
Citations
7,255
H-index
30
i10-index
43
Publications
20
Co-authors
40
list available
Publications
Resume (English only)
Academic Achievements
- VL-RewardBench: A Challenging Benchmark for Vision-Language Generative Reward Models, CVPR 2025 (Highlight, Top 3%)
- Video-MME: The First-Ever Comprehensive Evaluation Benchmark of Multi-modal LLMs in Video Analysis, CVPR 2025 (Highlight, Top 3%, Most Influential Paper of CVPR 2025)
- Temporal Reasoning Transfer from Text to Video, ICLR 2025
- VLFeedback: A Large-Scale AI Feedback Dataset for Large Vision-Language Models Alignment, EMNLP 2024
- VITATECS: A Diagnostic Dataset for Temporal Concept Understanding of Video-Language Models, ECCV 2024
- Multimodal ArXiv: A Dataset for Improving Scientific Comprehension of Large Vision-Language Models, ACL 2024
- Can Language Models Understand Physical Concepts?, EMNLP 2023
- M3IT: A Large-Scale Dataset towards Multi-Modal Multilingual Instruction Tuning, arxiv
- A Survey for In-context Learning, EMNLP 2024 (Most Influential Papers of EMNLP 2024)
Background
- Research interests include: developing frontier multimodal large language models (e.g., MiMo-VL, Reka Flash); understanding the fundamental mechanisms of LLMs and MLLMs (e.g., In-context Learning, LLMs-as-a-Judge). Currently a PhD student in the HKU-NLP group, co-supervised by Prof. Lingpeng Kong and Prof. Qi Liu. Master's degree from Peking University, advised by Prof. Xu Sun; Bachelor's degree from Xidian University.
Miscellany
- Always happy to discuss potential collaborations—feel free to reach out!
