Few-Shot Class-Incremental Image-to-Text on Food101
71.59AccuracyCoACT
Evaluation Results
| Method | Links | |
|---|---|---|
| CoACTBackbone=ViT-B/162024.05 | 71.59 | |
| LoRA2024.05 | 71.24 | |
| Linear Tuning2024.05 | 71.18 | |
| Prototype Learning2024.05 | 71.09 | |
| PriViLege2024.05 | 70.67 | |
| CPE-CLIP2024.05 | 69.71 |