Class-Incremental Learning on CIFAR-100 Uniformly Mild
2.01Average ForgettingCODAPrompt
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| CODAPromptScenario=Uniformly Mild, Backbone=ViT-B/16, Pre-trained=ImageNet-21k fine-tuned on ImageNet-1K2026.01 | 2.01 | 1 | 3.17 | |
| PROTEUSScenario=Uniformly Mild, Backbone=ViT-B/16, Pre-trained=ImageNet-21k fine-tuned on ImageNet-1K, LoRA rank r=42026.01 | 2.02 | 2 | 2.17 | |
| RanPACScenario=Uniformly Mild, Backbone=ViT-B/16, Pre-trained=ImageNet-21k fine-tuned on ImageNet-1K2026.01 | 2.87 | 3 | 2.5 | |
| SPromptScenario=Uniformly Mild, Backbone=ViT-B/16, Pre-trained=ImageNet-21k fine-tuned on ImageNet-1K2026.01 | 5.33 | 4 | 2.67 | |
| DualPromptScenario=Uniformly Mild, Backbone=ViT-B/16, Pre-trained=ImageNet-21k fine-tuned on ImageNet-1K2026.01 | 6.14 | 5 | 4.83 | |
| L2PScenario=Uniformly Mild, Backbone=ViT-B/16, Pre-trained=ImageNet-21k fine-tuned on ImageNet-1K2026.01 | 6.22 | 6 | 5.67 |