Text-to-Image Retrieval on DOCCI (test)
79.04R@1GOAL
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| GOALBackbone=ViT-L/142026.05 | 79.04 | 95.78 | 99.55 | 99.84 | |
| Long-CLIPBackbone=ViT-L/142026.05 | 78.84 | 95.25 | 99.19 | 99.59 | |
| FAST-GOALBackbone=ViT-L/142026.05 | 78.55 | 95.86 | 99.51 | 99.8 | |
| FAST-GOALBackbone=ViT-B/16, Zero-shot=true2026.05 | 74.27 | 93.76 | 99.27 | 99.71 | |
| EVA-CLIPBackbone=ViT-L/142026.05 | 73.08 | 93.1 | 98.25 | 99.27 | |
| Long-CLIPBackbone=ViT-B/16, Zero-shot=true2026.05 | 71.63 | 92.16 | 98.9 | 99.73 | |
| GOALBackbone=ViT-B/16, Zero-shot=true2026.05 | 71.22 | 92.39 | 98.9 | 99.61 | |
| EVA-CLIPBackbone=ViT-B/16, Zero-shot=true2026.05 | 66.61 | 89.08 | 97.53 | 99 | |
| FineCLIPBackbone=ViT-L/142026.05 | 66.29 | 88.37 | 97.24 | 98.75 | |
| FineCLIPBackbone=ViT-B/16, Zero-shot=true2026.05 | 62.92 | 87.22 | 96.67 | 98.35 | |
| CLIPBackbone=ViT-L/142026.05 | 60.51 | 86.31 | 96.67 | 98.45 | |
| CLIPBackbone=ViT-B/16, Zero-shot=true2026.05 | 58.53 | 85.57 | 97.16 | 98.82 |