Selective Classification on ImageNet-C Original (test)
58.3Metric AMDS
Evaluation results
| Method | Links | ||
|---|---|---|---|
| MDSBackbone=ResNet-50, Training Dataset=ImageNet-1K, Pre-training Source=Official PyTorch Repository2025.05 | 58.3 | 1.23 | |
| KNNBackbone=ResNet-50, Training Dataset=ImageNet-1K, Pre-training Source=Official PyTorch Repository2025.05 | 46.2 | 0.9 | |
| SIRCBackbone=ResNet-50, Training Dataset=ImageNet-1K, Pre-training Source=Official PyTorch Repository2025.05 | 40.5 | 0.729 | |
| Δ-KNNBackbone=ResNet-50, Training Dataset=ImageNet-1K, Pre-training Source=Official PyTorch Repository2025.05 | 33.9 | 0.529 | |
| EnergyBackbone=ResNet-50, Training Dataset=ImageNet-1K, Pre-training Source=Official PyTorch Repository2025.05 | 27.8 | 0.358 | |
| Δ-MDSBackbone=ResNet-50, Training Dataset=ImageNet-1K, Pre-training Source=Official PyTorch Repository2025.05 | 27.3 | 0.338 | |
| MaxLogitBackbone=ResNet-50, Training Dataset=ImageNet-1K, Pre-training Source=Official PyTorch Repository2025.05 | 24.5 | 0.26 | |
| MSPBackbone=ResNet-50, Training Dataset=ImageNet-1K, Pre-training Source=Official PyTorch Repository2025.05 | 22.6 | 0.203 | |
| RLogBackbone=ResNet-50, Training Dataset=ImageNet-1K, Pre-training Source=Official PyTorch Repository2025.05 | 22.4 | 0.193 | |
| Δ-KNN-RLogBackbone=ResNet-50, Training Dataset=ImageNet-1K, Pre-training Source=Official PyTorch Repository2025.05 | 22.2 | 0.189 | |
| Δ-MDS-RLogBackbone=ResNet-50, Training Dataset=ImageNet-1K, Pre-training Source=Official PyTorch Repository2025.05 | 21.9 | 0.18 |