Image Classification on CIFAR-100 5-shot (test)
85.2Top-1 AccEVA
Evaluation Results
| Method | Links | |
|---|---|---|
| EVAModel=BEIT-L/14, Pre-training Dataset=CLIP OpenAI, IN-22K, IN-1K, Evaluation Protocol=Linear Probing2024.11 | 85.2 | |
| SABackbone=ViT-Large, Shot-count=5-shot2025.03 | 81.98 | |
| Split LearningBackbone=ViT-Large, Shot-count=5-shot2025.03 | 79.63 | |
| Linear ProbingBackbone=ViT-Large, Shot-count=5-shot2025.03 | 78.79 | |
| SABackbone=Swin-L2025.03 | 76.66 | |
| Linear ProbingBackbone=Swin-L2025.03 | 75.76 | |
| Fine TuningBackbone=Swin-L2025.03 | 73.05 | |
| Offsite TuningBackbone=Swin-L2025.03 | 72.8 | |
| CLIPModel=ViT-B/16, Pre-training Dataset=LAION-300M, Evaluation Protocol=Linear Probing2024.11 | 71.96 | |
| ScaleKDModel=ViT-B/16, Pre-training Dataset=IN-1K, Evaluation Protocol=Linear Probing2024.11 | 70.91 | |
| CLIPModel=ViT-B/16, Pre-training Dataset=CLIP OpenAI, IN-12K, IN-1K, Evaluation Protocol=Linear Probing2024.11 | 70.85 | |
| CLIPModel=ViT-B/16, Pre-training Dataset=LAION-2B, IN-12K, IN-1K, Evaluation Protocol=Linear Probing2024.11 | 70.19 | |
| CLIPModel=ViT-B/16, Pre-training Dataset=LAION-2B, IN-1K, Evaluation Protocol=Linear Probing2024.11 | 69 | |
| Split LearningBackbone=Swin-L2025.03 | 65.37 | |
| Offsite TuningBackbone=ViT-Large, Shot-count=5-shot2025.03 | 64.83 | |
| Fine TuningBackbone=ViT-Large, Shot-count=5-shot2025.03 | 64.29 | |
| From-scratchModel=ViT-B/16, Pre-training Dataset=IN-1K, Evaluation Protocol=Linear Probing2024.11 | 60.3 | |
| LN-TUNEBackbone=ViT-Large, Shot-count=5-shot2025.03 | 34.8 | |
| LN-TUNEBackbone=Swin-L2025.03 | 29.23 |