Scholar
Chunyuan Li
Google Scholar ID: Zd7WmXUAAAAJ
xAI
Citations & Impact
Citations
47,672
H-index
72
i10-index
136
Publications
20
Co-authors
16
list available
Publications
Resume (English only)
Academic Achievements
- LLaVA series: including LLaVA-1.5 (SoTA on 11 open-source VLM benchmarks), LLaVA-NeXT, LLaVA-OneVision, LLaVA-Video, LLaVA-Critic, LLaVA-Med (NeurIPS 2023 Datasets and Benchmarks Track Spotlight), LLaVA-Interactive, and LLaVA-Plus.
- Developed the proprietary industry-leading VLM Seed-VL-1.5 for image and video understanding.
- Published numerous high-impact papers at NeurIPS (Oral/Spotlight), CVPR (Highlights), ECCV, and a survey in Foundations and Trends® in Computer Graphics and Vision.
- Notable projects include REACT (CVPR 2023), GLIGEN (CVPR 2023), X-Decoder, K-LITE (NeurIPS 2022 Oral), ELEVATER, and FocalNet.
- Authored a 110-page perspective paper 'Multimodal Foundation Models: From Specialists to General-Purpose Assistants' and delivered the CVPR 2023 tutorial on the topic.
- Served as Area Chair for NeurIPS, ICML, ICLR, EMNLP, TMLR, and Guest Editor for an IJCV special issue on 'the promises and dangers of large vision models'.
