AI Paper Atlas

Research map / Vision-language and multimodal / CLIP and zero-shot

GRIT-VLP: Grouped Mini-Batch Sampling for Efficient Vision and Language Pre-training

Jaeseok Byun, Taebaek Hwang, Jianlong Fu, Taesup Moon

ECCV 2022 · Vision-language and multimodal › CLIP and zero-shot · approach: Self-distillation for vision-language pretraining

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 hard instead of random in-batch negatives for ITM (no ITC) (ablation)ALBEF_Base (10 epochs) · MS-COCO TR R@1R@1: 61.6 -> 68.6Improved
add ITC to ITM_rand (ablation)ALBEF_Base (10 epochs) · MS-COCO TR R@1R@1: 61.6 -> 66.8Improved
add ITC to ITM_hard (ablation)ALBEF_Base (10 epochs) · MS-COCO TR R@1R@1: 68.6 -> 72.3Improved
raise MLM masking probability from 15% to 50% (ablation)ALBEF_Base (20 epochs) · MS-COCO TR R@1R@1: 73.8 -> 75.6Improved
raise MLM masking probability from 15% to 50% (ablation)ALBEF_Base (20 epochs) · NLVR2 testaccuracy: 80.37 -> 80.54No clear effect
add GRIT with shuffling to ALBEF_Base50 (ablation)ALBEF_Base50 · MS-COCO TR R@1R@1: 75.6 -> 76.4Improved
add GRIT with shuffling to ALBEF_Base50 (ablation)ALBEF_Base50 · NLVR2 testaccuracy: 80.54 -> 81.32Improved
remove both shufflings from GRIT (ablation)GRIT · MS-COCO TR R@1R@1: 76.4 -> 74.7Hurt
use naive GRIT instead of efficient GRIT (ablation)GRIT · epoch timetime per epoch: 2h 30m -> 3hHurt
add ITC consistency loss to GRIT at mask 50 (ablation)GRIT-VLP · MS-COCO TR R@1R@1: 76.4 -> 77.1Improved
add GRIT to ITC-only small model (ablation)ITC dual encoder · MS-COCO TR R@1R@1: 59.7 -> 63.0Improved
use queue-based ITC instead of plain ITC (ablation)ITC dual encoder · MS-COCO TR R@1R@1: 59.7 -> 60.3No clear effect

Later papers that compared with this method

2 later accepted papers report GRIT-VLP as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.

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