AI Paper Atlas

Research map / Detection and segmentation / CNN features

Understanding The Robustness in Vision Transformers

Daquan Zhou, Zhiding Yu, Enze Xie, Chaowei Xiao, Animashree Anandkumar, Jiashi Feng, Jose M. Alvarez

ICML 2022 · Detection and segmentation › CNN features · approach: Plain and high-resolution ViTs · 33 citations (OpenAlex)

This paper tried a common change: More attention heads

Increase the number of attention heads · ImageNet robustness · robustness Mixed

What 107 accepted papers report for this change

See this change in detailMatched by the wording of each paper's own result, read by AI from the full text. Papers report gains more often than failures, so this shows what gets published, not the chance a change works.

Read the paper

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 FAN-ECA channel processing instead of SE channel attention (ablation)FAN-ViT-S · ImageNet-CIN-C accuracy: 62.0 -> 64.6Improved
Use FAN-ECA channel processing instead of self-attention channel processing (ablation)FAN-ViT-S · ImageNet-CIN-C accuracy: 61.7 -> 64.6Improved
Replace conventional transformer blocks with FAN blocks (ablation)ViT-S* · ImageNet-CIN-C accuracy: 58.1 -> 61.7Improved
Add FAN blocks to shifted-window Swin-T (ablation)Swin-T · ImageNet-CIN-C accuracy: 55.4 -> 59.4Improved
Add FAN to ConvNeXt-T (ablation)ConvNeXt-T · ImageNet-CIN-C accuracy: 59.1 -> 60.8Improved
Add the DeiT training recipe (ablation)ViT-S · ImageNet-CIN-C accuracy: 54.2 -> 57.1Improved
Increase ViT-S depth from 8 to 12 blocks (ablation)ViT-S with DeiT recipe · ImageNet-CIN-C accuracy: 57.1 -> 58.0Improved
Add knowledge distillation (ablation)ViT-S with 12 blocks · ImageNet-CIN-C accuracy: 58.0 -> 59.6Improved
Use ImageNet-22K pretraining without KD (ablation)ViT-S with 12 blocks · ImageNet-CIN-C accuracy: 58.0 -> 59.7Improved
Add the DeiT training recipe (ablation)ResNet-50 · ImageNet-CIN-C accuracy: 38.8 -> 43.9Improved
Add SE channel attention (ablation)ResNet-50 with DeiT recipe · ImageNet-CIN-C accuracy: 43.9 -> 50.1Improved
Use strided convolution for downsampling (ablation)ResNet-50 with DeiT recipe and SE · ImageNet-CIN-C accuracy: 50.1 -> 52.1Improved

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