DnA: Denoising Attention for Visual Tasks

Open paper
Benchmark Model Rank Results
image-classification-on-cifar-100DnA (ViT-B)#49Percentage correct: 89.5
image-classification-on-cifar-10DnA (ViT-B)#50Percentage correct: 98.6
image-classification-on-imagenetDnA (ViT-B)#637Top 1 Accuracy: 81.9%

Rank counts only results with a code link.