Image Classification on CIFAR-10 Bucket 1 Forget
98.2AccuracyOriginal
Evaluation results
| Method | Links | |
|---|---|---|
| OriginalEvaluation Head=Output, Backbone=ViT-S/16, Pre-training=ImageNet2026.04 | 98.2 | |
| OriginalEvaluation Head=Linear Probe, Backbone=ViT-S/16, Pre-training=ImageNet2026.04 | 98.17 | |
| OriginalEvaluation Head=NCC, Backbone=ViT-S/16, Pre-training=ImageNet2026.04 | 98.06 | |
| Retain-only RetainEvaluation Head=Linear Probe, Backbone=ViT-S/16, Pre-training=ImageNet2026.04 | 97.71 | |
| Retain-only RetainEvaluation Head=NCC, Backbone=ViT-S/16, Pre-training=ImageNet2026.04 | 94.24 | |
| Random-label with CMFEvaluation Head=Linear Probe, Backbone=ViT-S/16, Pre-training=ImageNet2026.04 | 72.3 | |
| Salun with CMFEvaluation Head=Linear Probe, Backbone=ViT-S/16, Pre-training=ImageNet2026.04 | 71.98 | |
| Salun with CMFEvaluation Head=NCC, Backbone=ViT-S/16, Pre-training=ImageNet2026.04 | 56.14 | |
| Random-label with CMFEvaluation Head=NCC, Backbone=ViT-S/16, Pre-training=ImageNet2026.04 | 55.82 | |
| NegGrad+ with CMFEvaluation Head=Linear Probe, Backbone=ViT-S/16, Pre-training=ImageNet2026.04 | 55.61 | |
| Salun with CMFEvaluation Head=Output, Backbone=ViT-S/16, Pre-training=ImageNet2026.04 | 53.69 | |
| Random-label with CMFEvaluation Head=Output, Backbone=ViT-S/16, Pre-training=ImageNet2026.04 | 52.77 | |
| NegGrad+ with CMFEvaluation Head=NCC, Backbone=ViT-S/16, Pre-training=ImageNet2026.04 | 43.89 | |
| NegGrad+ with CMFEvaluation Head=Output, Backbone=ViT-S/16, Pre-training=ImageNet2026.04 | 41.42 | |
| Retain-only RetainEvaluation Head=Output, Backbone=ViT-S/16, Pre-training=ImageNet2026.04 | 0 |