AI Paper Atlas

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

COOKIE: Contrastive Cross-Modal Knowledge Sharing Pre-Training for Vision-Language Representation

Keyu Wen, Jin Xia, Yuanyuan Huang, Linyang Li, Jiayan Xu, Jie Shao

ICCV 2021 · Vision-language and multimodal › Cross-modal learning · approach: Probabilistic embeddings for cross-modal retrieval · 37 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 cross-modal contrastive pre-training (ablation)COOKIE · MSCOCO 1K test ITMRsum: 526.9 -> 545.4Improved
add visual contrastive learning on top of CCL (ablation)COOKIE · MSCOCO 1K test ITMRsum: 545.4 -> 547.8Improved
add textual contrastive learning on top of CCL (ablation)COOKIE · STS-B text matchingPearson/Spearman mean: 72.4 -> 75.1Improved
add visual contrastive learning on top of CCL (ablation)COOKIE · STS-B text matchingPearson/Spearman mean: 72.4 -> 72.1No clear effect
add weight-sharing transformer encoder (ablation)COOKIE · MSCOCO 1K test ITMRsum: 536.6 -> 547.5Improved
share weights in WS-TE between image and text (ablation)COOKIE · MSCOCO 1K test ITMRsum: 541.6 -> 547.5Improved
add text-aligned visual transformer (TAV-TE) with weight sharing (ablation)COOKIE · MSCOCO 1K test ITMRsum: 547.5 -> 550.0Improved
replace WS-TE with FC layers (ablation)COOKIE · MSCOCO 1K test ITMRsum: 536.6 -> 536.8No clear effect
use 2 WS-TE layers instead of 1 with pre-training (ablation)COOKIE · MSCOCO 1K test ITMRsum: 547.8 -> 550.0Improved
use 8 WS-TE layers instead of 2 without pre-training (ablation)COOKIE · MSCOCO 1K test ITMRsum: 526.9 -> 250.8Hurt
increase pre-training corpus from CC 2.8M to 5.9M pairs (ablation)COOKIE · MSCOCO 1K test ITMRsum: 540.5 -> 550.0Improved
add VQA and GQA to pre-training corpus (ablation)COOKIE · MSCOCO 1K test ITMRsum: 547.6 -> 550.0Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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