Image Classification on CIFAR-100 with Asymmetric Noise (1k clean samples)
71.53Accuracy (eta=0.2)MW-Net
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| MW-NetModel Selection=Best2026.03 | 71.53 | 67.64 | 58.62 | |
| Cross-EntropyModel Selection=Best2026.03 | 70.57 | 66.48 | 57.17 | |
| Robust Self-Training with Closed-loop Label CorrectionModel Selection=Best2026.03 | 69.92 | 66.99 | 58.58 | |
| Robust Self-Training with Closed-loop Label CorrectionModel Selection=Last2026.03 | 67.17 | 60.52 | 53.62 | |
| FasTENModel Selection=Best2026.03 | 64.79 | 60.77 | 60.79 | |
| MW-NetModel Selection=Last2026.03 | 64.3 | 49.91 | 32.16 | |
| L2BModel Selection=Best2026.03 | 63.8 | 57.83 | 42.85 | |
| Cross-EntropyModel Selection=Last2026.03 | 63.6 | 48.58 | 32.32 | |
| L2BModel Selection=Last2026.03 | 63.56 | 57.53 | 42.83 | |
| FasTENModel Selection=Last2026.03 | 58.1 | 46.09 | 45.32 |