Image Classification on ImageNet-1K (val) (Adversarial Robustness Profile)
84.3Clean AccuracyTAPADL-FAN
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| TAPADL-FANParams (M)=50.7, FLOPS (G)=11.8, Pre-training=ImageNet-1K2024.06 | 84.3 | 18.6 | 9.2 | 13.5 | 16.9 | |
| FAN-B-AdaNCAParams (M)=51.7, FLOPS (G)=12.4, Pre-training=ImageNet-1K2024.06 | 84.1 | 20.3 | 10.6 | 14.1 | 19.1 | |
| FAN-BParams (M)=50.4, FLOPS (G)=11.7, Pre-training=ImageNet-1K2024.06 | 83.9 | 15 | 7.6 | 10.4 | 13.1 | |
| Swin-B-AdaNCAParams (M)=90.7, FLOPS (G)=16.3, Pre-training=ImageNet-1K2024.06 | 83.7 | 24.1 | 20.5 | 25.1 | 24.8 | |
| Swin-BParams (M)=87.8, FLOPS (G)=15.4, Pre-training=ImageNet-1K2024.06 | 83.4 | 21.3 | 13.4 | 15.6 | 23.1 | |
| RVT-B-AdaNCAParams (M)=91, FLOPS (G)=19, Pre-training=ImageNet-1K2024.06 | 83.3 | 36.7 | 30.2 | 33.2 | 36.2 | |
| Swin-BParams (M)=94.1, FLOPS (G)=16.7, Pre-training=ImageNet-22K2024.06 | 83.3 | 22.8 | 14.6 | 15.9 | 23.8 | |
| ConViT-B-AdaNCAParams (M)=89, FLOPS (G)=19, Pre-training=ImageNet-1K2024.06 | 83.2 | 29.2 | 20.1 | 26.3 | 28.4 | |
| TAPADL-RVTParams (M)=89.4, FLOPS (G)=17.9, Pre-training=ImageNet-1K2024.06 | 83.1 | 27.6 | 19.3 | 17.7 | 26.8 | |
| RVT-BParams (M)=88.5, FLOPS (G)=17.7, Pre-training=ImageNet-1K2024.06 | 82.7 | 29.9 | 21.5 | 21.9 | 31.4 | |
| ConViT-BParams (M)=93.6, FLOPS (G)=19.2, Pre-training=ImageNet-22K2024.06 | 82.7 | 24.1 | 10 | 20.5 | 23.9 | |
| ConViT-BParams (M)=86.5, FLOPS (G)=17.7, Pre-training=ImageNet-1K2024.06 | 82.4 | 21.2 | 8.9 | 16.9 | 20.3 |