Few-Shot Class-Incremental Image-to-Text on CUB200
77.38AccuracyCoACT
Evaluation Results
| Method | Links | |
|---|---|---|
| CoACTBackbone=ViT-B/162024.05 | 77.38 | |
| LoRA2024.05 | 74.65 | |
| PriViLege2024.05 | 74.45 | |
| Linear Tuning2024.05 | 74.41 | |
| Prototype Learning2024.05 | 74.1 | |
| CPE-CLIP2024.05 | 73.47 |