Cross-modal Retrieval on Flickr
97.2TR R@5RG-TTA
Evaluation results
| Method | Links | ||
|---|---|---|---|
| RG-TTABackbone=ViT-B/16, Zero-shot=true2026.07 | 97.2 | 88.4 | |
| RG-TTABackbone=ViT-B/32, Zero-shot=true2026.07 | 97 | 86.5 | |
| Joint V-LBackbone=ViT-B/16, Zero-shot=true2026.07 | 96.5 | 83.8 | |
| Original CLIPBackbone=ViT-B/16, Zero-shot=true2026.07 | 96.4 | 85.5 | |
| CLIP-clipBackbone=ViT-B/16, Zero-shot=true2026.07 | 95.4 | 85.3 | |
| Original CLIPBackbone=ViT-B/32, Zero-shot=true2026.07 | 94.7 | 83.5 | |
| Joint V-LBackbone=ViT-B/32, Zero-shot=true2026.07 | 94.2 | 83 | |
| Biased-promptsBackbone=ViT-B/16, Zero-shot=true2026.07 | 94.1 | 85.8 | |
| CLIP-clipBackbone=ViT-B/32, Zero-shot=true2026.07 | 92.6 | 81.2 | |
| Biased-promptsBackbone=ViT-B/32, Zero-shot=true2026.07 | 91.9 | 83.3 |