Image Classification on CIFAR-10 Unfiltered (test)
90.52AccuracyClean
Evaluation results
| Method | Links | ||
|---|---|---|---|
| CleanBackbone=ResNet50×4, Evaluation Protocol=Fine-tune (last block + pooling)2026.06 | 90.52 | 0 | |
| TUEBackbone=ResNet50×4, Evaluation Protocol=Fine-tune (last block + pooling)2026.06 | 90.49 | 0.03 | |
| LSPBackbone=ResNet50×4, Evaluation Protocol=Fine-tune (last block + pooling)2026.06 | 90.44 | 0.08 | |
| EMNBackbone=ResNet50×4, Evaluation Protocol=Fine-tune (last block + pooling)2026.06 | 90.19 | 0.33 | |
| GUEBackbone=ResNet50×4, Evaluation Protocol=Fine-tune (last block + pooling)2026.06 | 89.69 | 0.83 | |
| PUEBackbone=ResNet50×4, Evaluation Protocol=Fine-tune (last block + pooling)2026.06 | 89.63 | 0.89 | |
| FUSEBackbone=ResNet50×4, Evaluation Protocol=Fine-tune (last block + pooling)2026.06 | 86.24 | 4.28 | |
| Zero-shotBackbone=ResNet50×4, Evaluation Protocol=Zero-shot2026.06 | 79.21 | — |