Scholar
Xiaohan Wang
Google Scholar ID: iGA10XoAAAAJ
Stanford University
Citations & Impact
Citations
2,569
H-index
28
i10-index
41
Publications
20
Co-authors
0
Publications
Resume
Academic Achievements
- Publications:
- - Temporal Preference Optimization for Long-Form Video Understanding (2025)
- - Apollo: An Exploration of Video Understanding in Large Multimodal Models (2024)
- - Video-STaR: Bootstrapping Weak Video Supervision for Visual Instruction Tuning (2025)
- - Video Action Differencing (2025)
- - Feather the Throttle: Revisiting Visual Token Pruning for Vision-Language Model Acceleration (2024)
- Talks and Project Releases:
- - Gave a talk at 'What is Next in Video Understanding' workshop @ CVPR 2024
- - Released Temporal Preference Optimization (TPO) framework
- - Released Apollo project
- - VLM Classifier accepted to NeurIPS 2024
- - VideoAgent accepted to ECCV 2024
- - VisDiff accepted as an oral presentation at CVPR 2024
- - RLCF accepted by ICLR 2024
Research Experience
- Collaborated with researchers at Baidu Research and Facebook AI Research during Ph.D. studies. Currently working at Stanford University with Prof. Serena Yeung.
Education
- Ph.D.: University of Technology Sydney, advised by Prof. Yi Yang; B.E.: University of Science and Technology of China.
Background
- Research interests include Video Understanding, Multimodal Learning, and AI for Healthcare. Currently a Postdoc at Stanford University, affiliated with MARVL and Stanford AI Lab.