Image Classification on ImageNet Clean 1K (val)
89.71Top-1 AccKoo-Fu CLIP
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Koo-Fu CLIPf(·) (Prototype transformation)=1K (Sw:1K), class. (Search Space)=1K2026.02 | 89.71 | — | 98.19 | |
| Koo-Fu CLIPf(·) (Prototype transformation)=21K (Sw:21K), class. (Search Space)=1K2026.02 | 88.11 | — | 98.03 | |
| CLIPf(·) (Prototype transformation)=None (-), class. (Search Space)=1K2026.02 | 85.82 | — | 96.84 | |
| FastViT-MA36Number of Parameters=42.7M, FLOPs=7.9G2023.03 | 83.9 | — | — | |
| FastViT-SA36Number of Parameters=30.4M, FLOPs=5.6G2023.03 | 83.6 | — | — | |
| ConvNeXt-SNumber of Parameters=50.0M, FLOPs=8.7G2023.03 | 83.1 | — | — | |
| EfficientNet-B4Number of Parameters=19.3M, FLOPs=4.2G2023.03 | 83 | — | — | |
| FastViT-SA24Number of Parameters=20.6M, FLOPs=3.8G2023.03 | 82.6 | — | — | |
| ConvNeXt-TNumber of Parameters=28.6M, FLOPs=4.5G2023.03 | 82.1 | — | — | |
| RVT-SNumber of Parameters=22.1M, FLOPs=4.7G2023.03 | 81.7 | — | — | |
| ConViT-SNumber of Parameters=27.8M, FLOPs=5.4G2023.03 | 81.5 | — | — | |
| Swin-TNumber of Parameters=28.3M, FLOPs=4.5G2023.03 | 81.2 | — | — | |
| FastViT-SA12Number of Parameters=10.9M, FLOPs=1.9G2023.03 | 80.6 | — | — | |
| PVT-SmallNumber of Parameters=24.5M, FLOPs=3.8G2023.03 | 79.9 | — | — | |
| ResNeXt50-32x4dNumber of Parameters=25.0M, FLOPs=4.3G2023.03 | 79.8 | — | — | |
| ResNet-50Number of Parameters=25.6M, FLOPs=4.1G, AugMix=true2023.03 | 79 | — | — | |
| RVT-TiNumber of Parameters=8.6M, FLOPs=1.3G2023.03 | 78.4 | — | — | |
| PVT-TinyNumber of Parameters=13.2M, FLOPs=1.9G2023.03 | 75 | — | — | |
| Soft-ACMoEBackbone=SoftMoE2025.02 | 73.21 | — | 91.23 | |
| SoftMoEBackbone=SoftMoE2025.02 | 72.86 | — | 90.92 | |
| Koo-Fu CLIPf(·) (Prototype transformation)=1K (Sw:1K), class. (Search Space)=14K2026.02 | 68.28 | — | 74.15 | |
| Koo-Fu CLIPf(·) (Prototype transformation)=21K (Sw:21K), class. (Search Space)=14K2026.02 | 67.96 | — | 74.08 | |
| CLIPf(·) (Prototype transformation)=None (-), class. (Search Space)=14K2026.02 | 65.76 | — | 73.27 | |
| Koo-Fu CLIPf(·) (Prototype transformation)=1K (Sw:1K), class. (Search Space)=21K2026.02 | 55.37 | — | 84.25 | |
| Koo-Fu CLIPf(·) (Prototype transformation)=21K (Sw:21K), class. (Search Space)=21K2026.02 | 53.71 | — | 83.41 | |
| CLIPf(·) (Prototype transformation)=None (-), class. (Search Space)=21K2026.02 | 50.96 | — | 80.15 | |
| ASAM#Epochs=100, Runtime=41.1h, Backbone=ViT-S16, Data Augmentation=Basic Inception-style2024.06 | — | 24.01 | — | |
| ASAM#Epochs=300, Runtime=123h, Backbone=ViT-S16, Data Augmentation=MixUp and RandAugment2024.06 | — | 20.38 | — | |
| ASAM#Epochs=150, Runtime=136h, Backbone=ViT-B16, Data Augmentation=MixUp and RandAugment2024.06 | — | 19.4 | — | |
| DAMP#Epochs=100, Runtime=20.7h, Backbone=ViT-S16, Data Augmentation=Basic Inception-style2024.06 | — | 25.5 | — | |
| DAMP#Epochs=200, Runtime=41.1h, Backbone=ViT-S16, Data Augmentation=Basic Inception-style2024.06 | — | 23.75 | — | |
| DAMP#Epochs=500, Runtime=111h, Backbone=ViT-S16, Data Augmentation=MixUp and RandAugment2024.06 | — | 20.09 | — | |
| DAMP#Epochs=275, Runtime=124h, Backbone=ViT-B16, Data Augmentation=MixUp and RandAugment2024.06 | — | 19.36 | — | |
| Dropout#Epochs=100, Runtime=20.6h, Backbone=ViT-S16, Data Augmentation=Basic Inception-style2024.06 | — | 28.55 | — | |
| Dropout#Epochs=200, Runtime=41.1h, Backbone=ViT-S16, Data Augmentation=Basic Inception-style2024.06 | — | 28.74 | — | |
| Dropout#Epochs=500, Runtime=111h, Backbone=ViT-S16, Data Augmentation=MixUp and RandAugment2024.06 | — | 20.25 | — | |
| Dropout#Epochs=275, Runtime=123h, Backbone=ViT-B16, Data Augmentation=MixUp and RandAugment2024.06 | — | 20.41 | — | |
| SAM#Epochs=100, Runtime=41h, Backbone=ViT-S16, Data Augmentation=Basic Inception-style2024.06 | — | 23.91 | — | |
| SAM#Epochs=300, Runtime=123h, Backbone=ViT-S16, Data Augmentation=MixUp and RandAugment2024.06 | — | 20.17 | — | |
| SAM#Epochs=150, Runtime=135h, Backbone=ViT-B16, Data Augmentation=MixUp and RandAugment2024.06 | — | 19.84 | — |