AI Paper Atlas

Research map / Architectures and efficiency / Attention and transformers

DeiT III: Revenge of the ViT

Hugo Touvron, Matthieu Cord, Hervé Jégou

ECCV 2022 · Architectures and efficiency › Attention and transformers · approach: Hybrid and interpretable ViT designs · 520 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 binary cross-entropy instead of cross-entropy (ablation)ViT-B · ImageNet-1k valtop-1 accuracy: 80.9 -> 82.2Improved
use binary cross-entropy instead of cross-entropy (ablation)ViT-S · ImageNet-1k valtop-1 accuracy: 79.8 -> 79.8No clear effect
add LayerScale (ablation)ViT-B · ImageNet-1k valtop-1 accuracy: 82.2 -> 82.5Improved
add LayerScale (ablation)ViT-S · ImageNet-1k valtop-1 accuracy: 79.8 -> 80.1Improved
replace RandAugment with 3-Augment (ablation)ViT-B · ImageNet-1k valtop-1 accuracy: 82.7 -> 83.1Improved
replace RandAugment with 3-Augment (ablation)ViT-L · ImageNet-1k valtop-1 accuracy: 84.0 -> 84.2Improved
replace Auto-Augment with 3-Augment on ResNet-50 (ablation)ResNet-50 · ImageNet-1k valtop-1 accuracy: 79.7 -> 79.4Hurt
add grayscale only to color jitter (ablation)ViT-B · ImageNet-1k valtop-1 accuracy: 81.4 -> 81.0Hurt
add solarization to jitter, grayscale and blur (ablation)ViT-B · ImageNet-1k valtop-1 accuracy: 82.7 -> 83.1Improved
add color jitter 0.3 to 3-Augment (ablation)ViT-B · ImageNet-1k valtop-1 accuracy: 83.1 -> 83.1No clear effect
use Simple Random Crop instead of Random Resized Crop (ablation)ViT-S · ImageNet-1k val (21k pretrain)top-1 accuracy: 81.6 -> 82.1Improved
use Simple Random Crop instead of Random Resized Crop (ablation)ViT-S · ImageNet-1k val (1k only)top-1 accuracy: -0.9%Hurt

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