Class-Incremental Learning on Cars196
86.9Final AccuracyJoint full tuning
Evaluation results
| Method | Links | ||
|---|---|---|---|
| Joint full tuningBackbone=ViT-B/16, Pre-training=ImageNet21K, Tasks=102024.11 | 86.9 | — | |
| SoTUBackbone=ViT-B/16, Pre-training=ImageNet21K, Tasks=102024.11 | 78.8 | 85.3 | |
| RanPACBackbone=ViT-B/16, Pre-training=ImageNet21K, Tasks=102024.11 | 75.1 | 82.8 | |
| DERBackbone=ViT-B/16, Pre-training=ImageNet21K, Tasks=10, Examples per class=202024.11 | 69.9 | 75.2 | |
| iCaRLBackbone=ViT-B/16, Pre-training=ImageNet21K, Tasks=10, Examples per class=202024.11 | 62.4 | 74.7 | |
| MEMOBackbone=ViT-B/16, Pre-training=ImageNet21K, Tasks=10, Examples per class=202024.11 | 60.5 | 70.6 | |
| ESNBackbone=ViT-B/16, Pre-training=ImageNet21K, Tasks=102024.11 | 56.9 | 72.8 | |
| HiDe-PromptBackbone=ViT-B/16, Pre-training=ImageNet21K, Tasks=102024.11 | 52 | 53.6 | |
| CODA-PromptBackbone=ViT-B/16, Pre-training=ImageNet21K, Tasks=102024.11 | 45.4 | 52.1 | |
| L2PBackbone=ViT-B/16, Pre-training=ImageNet21K, Tasks=102024.11 | 45.3 | 58.2 | |
| ADAM (Ada)Backbone=ViT-B/16, Pre-training=ImageNet21K, Tasks=10, variant=Ada2024.11 | 43.2 | 56 | |
| FOSTERBackbone=ViT-B/16, Pre-training=ImageNet21K, Tasks=10, Examples per class=202024.11 | 42.9 | 44.3 | |
| ADAM (SSF)Backbone=ViT-B/16, Pre-training=ImageNet21K, Tasks=10, variant=SSF2024.11 | 40.5 | 52.8 | |
| SimpleCILBackbone=ViT-B/16, Pre-training=ImageNet21K, Tasks=102024.11 | 37.8 | 50.4 | |
| DualPromptBackbone=ViT-B/16, Pre-training=ImageNet21K, Tasks=102024.11 | 37.7 | 53.4 |