What matters when building vision-language models?
Research paper by Hugo Laurençon and collaborators. Published 2024-05-03. Topics: vision-language models, large language models, vision transformers, pre-trained models, architecture choice.
Resource information
- Authors
- Hugo Laurençon, Léo Tronchon, Matthieu Cord, Victor Sanh
- Research topics
- vision-language models, large language models, vision transformers, pre-trained models, architecture choice, Idefics2, multimodal benchmarks
- Publication date
- 2024-05-03
- Publication
- arXiv
- Last record update
- Nov 3, 2026