| visual-prompt-tuning-on-fgvc | SPT-Deep(ViT-B/16_MAE_pretrained_ImageNet-1K) | – | Mean Accuracy: 83.26 |
| visual-prompt-tuning-on-fgvc | SPT-Deep(ViT-B/16_MoCo_v3_pretrained_ImageNet-1K) | – | Mean Accuracy: 86.00 |
| visual-prompt-tuning-on-fgvc | SPT-Shallow(ViT-B/16_MAE_pretrained_ImageNet-1K) | – | Mean Accuracy: 73.95 |
| visual-prompt-tuning-on-fgvc | SPT-Shallow(ViT-B/16_MoCo_v3_pretrained_ImageNet-1K) | – | Mean Accuracy: 84.08 |
| visual-prompt-tuning-on-vtab-1k-natural-7 | SPT-Deep(ViT-B/16_MAE_pretrained_ImageNet-1K) | – | Mean Accuracy: 67.19 |
| visual-prompt-tuning-on-vtab-1k-natural-7 | SPT-Deep(ViT-B/16_MoCo_v3_pretrained_ImageNet-1K) | – | Mean Accuracy: 76.20 |
| visual-prompt-tuning-on-vtab-1k-natural-7 | SPT-Shallow(ViT-B/16_MAE_pretrained_ImageNet-1K) | – | Mean Accuracy: 62.53 |
| visual-prompt-tuning-on-vtab-1k-natural-7 | SPT-Shallow(ViT-B/16_MoCo_v3_pretrained_ImageNet-1K) | – | Mean Accuracy: 74.47 |
| visual-prompt-tuning-on-vtab-1k-specialized-4 | SPT-Deep(ViT-B/16_MAE_pretrained_ImageNet-1K) | – | Mean Accuracy: 83.15 |
| visual-prompt-tuning-on-vtab-1k-specialized-4 | SPT-Deep(ViT-B/16_MoCo_v3_pretrained_ImageNet-1K) | – | Mean Accuracy: 84.95 |
| visual-prompt-tuning-on-vtab-1k-specialized-4 | SPT-Shallow(ViT-B/16_MAE_pretrained_ImageNet-1K) | – | Mean Accuracy: 80.90 |
| visual-prompt-tuning-on-vtab-1k-specialized-4 | SPT-Shallow(ViT-B/16_MoCo_v3_pretrained_ImageNet-1K) | – | Mean Accuracy: 83.93 |
| visual-prompt-tuning-on-vtab-1k-structured-8 | SPT-Deep(ViT-B/16_MAE_pretrained_ImageNet-1K) | – | Mean Accuracy: 59.23 |
| visual-prompt-tuning-on-vtab-1k-structured-8 | SPT-Deep(ViT-B/16_MoCo_v3_pretrained_ImageNet-1K) | – | Mean Accuracy: 58.36 |
| visual-prompt-tuning-on-vtab-1k-structured-8 | SPT-Shallow(ViT-B/16_MAE_pretrained_ImageNet-1K) | – | Mean Accuracy: 53.46 |
| visual-prompt-tuning-on-vtab-1k-structured-8 | SPT-Shallow(ViT-B/16_MoCo_v3_pretrained_ImageNet-1K) | – | Mean Accuracy: 55.16 |