AI Paper Atlas

Research map / Vision-language and multimodal / Cross-modal learning

Contrastive Vision-Language Pre-training with Limited Resources

Quan Cui, Boyan Zhou, Yu Guo, Weidong Yin, Hao Wu, Osamu Yoshie, Yubo Chen

ECCV 2022 · Vision-language and multimodal › Cross-modal learning · approach: Contrastive vision-language alignment · 30 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
add debiased sampling (same-dataset batches) (ablation)naive baseline · MS-COCO 5K zero-shotRSUM: 371.6 -> 392.3Improved
add debiased sampling (same-dataset batches) (ablation)naive baseline · Flickr30K 1K zero-shotRSUM: 483.3 -> 510.1Improved
add coin flipping mixup (ablation)baseline + debiased sampling · MS-COCO 5K zero-shotRSUM: 392.3 -> 402.8Improved
add coin flipping mixup (ablation)baseline + debiased sampling · Flickr30K 1K zero-shotRSUM: 510.1 -> 519.2Improved
add gradient reserved gather (ablation)baseline + data methods · MS-COCO 5K zero-shotRSUM: 402.8 -> 415.0Improved
add gradient reserved gather (ablation)baseline + data methods · Flickr30K 1K zero-shotRSUM: 519.2 -> 524.1Improved
increase batch size from 1,024 to 16,384 (ablation)ZeroVL · MS-COCO 5K zero-shotRSUM: 415.0 -> 425.0Improved
increase batch size from 1,024 to 16,384 (ablation)ZeroVL · Flickr30K 1K zero-shotRSUM: 524.1 -> 536.2Improved
use decoupled gradient accumulation (16 steps) on 8 GPUs instead of batch 16,384 on 128 GPUs (ablation)ZeroVL · MS-COCO 5K zero-shotRSUM: 425.0 vs 425.2No clear effect
add identical random seed for the two forwards in DGA (ablation)decoupled gradient accumulation · MS-COCO 5K zero-shotRSUM: 413.4 -> 425.2Improved
add identical random seed for the two forwards in DGA (ablation)decoupled gradient accumulation · Flickr30K 1K zero-shotRSUM: 527.1 -> 535.9Improved
add TokenDrop (mask 25% of ViT tokens) (ablation)ZeroVL with DGA · MS-COCO 5K zero-shotRSUM / GPU time: 425.2 -> 424.8; ~680 -> ~470 hrNo clear effect

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