Few-Shot Class-Incremental Learning on CIFAR-100 (60 base classes, 5-way 5-shot, Session 1)
89.39AccuracyPriViLege
Evaluation results
| Method | Links | |
|---|---|---|
| PriViLegePT. Backbone=CLIP-B/162024.05 | 89.39 | |
| CoACTPT. Backbone=ViT-B/162024.05 | 88.46 | |
| CPE-CLIPPT. Backbone=CLIP-B/162024.05 | 85.86 | |
| SV-TPT. Backbone=SwinT2024.05 | 82.82 | |
| BOTPT. Backbone=X2024.05 | 77.2 | |
| SoftNetPT. Backbone=X2024.05 | 75.54 | |
| SAVCPT. Backbone=X2024.05 | 73.7 | |
| C-FSCILPT. Backbone=X2024.05 | 72.4 | |
| LIMITPT. Backbone=X2024.05 | 72.09 | |
| FACTPT. Backbone=X2024.05 | 72.09 | |
| MetaFSCILPT. Backbone=X2024.05 | 70.1 | |
| CLOMPT. Backbone=X2024.05 | 69.83 | |
| CECPT. Backbone=X2024.05 | 68.88 | |
| F2MPT. Backbone=X2024.05 | 63.8 | |
| FSLL+SSPT. Backbone=X2024.05 | 63.14 | |
| FSLLPT. Backbone=X2024.05 | 61.75 | |
| IDLVQ-CPT. Backbone=X2024.05 | 59.87 | |
| TOPICPT. Backbone=X2024.05 | 50.09 | |
| RebalancingPT. Backbone=X2024.05 | 47.8 | |
| iCaRLPT. Backbone=X2024.05 | 46.32 |