← Search

Leo Tronchon

2 accepted papers

2024

What matters when building vision-language models?

NeurIPS 2024poster

The growing interest in vision-language models (VLMs) has been driven by improvements in large language models and vision transformers. Despite the abundance of literature on this subject, we observe that critical decisions regarding the design of VLMs are often not justified. We argue that these un…

Cited by 213SourcePDFScholar
2023

OBELICS: An Open Web-Scale Filtered Dataset of Interleaved Image-Text Documents

NeurIPS 2023poster

Large multimodal models trained on natural documents, which interleave images and text, outperform models trained on image-text pairs on various multimodal benchmarks. However, the datasets used to train these models have not been released, and the collection process has not been fully specified. W…