AI Paper Atlas

Research map / Vision-language and multimodal / Human and embodied modeling

Multimodal C4: An Open, Billion-scale Corpus of Images Interleaved with Text

Wanrong Zhu, Jack Hessel, Anas Awadalla, Samir Yitzhak Gadre, Jesse Dodge, Alex Fang, Youngjae Yu, Ludwig Schmidt, William Yang Wang, Yejin Choi

NeurIPS 2023 · Vision-language and multimodal › Human and embodied modeling · approach: Vision-language pretraining · 17 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Use zero-shot CLIP ViT-L/14 for image-sentence matchingWithin-document image-text alignment · Story-DIIAUC: 92.8 vs. 85.5Improved
Use zero-shot CLIP ViT-L/14 for image-sentence matchingWithin-document image-text alignment · Story-SISP@1: 73.3 vs. 53.1Improved
Use zero-shot CLIP ViT-L/14 for image-sentence matchingWithin-document image-text alignment · DII-StressAUC: 98.7 vs. 95.3Improved
Use zero-shot CLIP ViT-L/14 for image-sentence matchingWithin-document image-text alignment · RQAAUC: 80.7 vs. 69.3Improved
Use zero-shot CLIP ViT-L/14 for image-sentence matchingWithin-document image-text alignment · DIYAUC: 74.0 vs. 61.8Improved
Replace max-similarity matching with bipartite linear assignment (ablation)Within-document image-text assignment · mmc4 document sequencesSentences associated with an image: 22% -> 34%Improved
Replace max-similarity matching with bipartite linear assignment (ablation)Within-document image-text assignment · mmc4 document sequencesMean CLIP similarity: 24.5 -> 24.0Hurt
Train on interleaved mmc4-core sequences instead of image-caption pairsOpenFlamingo-3B · COCO image captioning, 4-shot and 8-shotCIDEr: 20-30 CIDEr point improvementsImproved
Train on image-caption pairs rather than interleaved sequences (ablation)OpenFlamingo-3B · COCO image captioning, 4-shot to 8-shotCIDErHurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy