Image-to-Text Retrieval on DCI (test)
60.76R@1FAST-GOAL
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| FAST-GOALBackbone=ViT-L/142026.05 | 60.76 | 78.93 | 90.42 | 93.61 | |
| FAST-GOALBackbone=ViT-B/16, Zero-shot=true2026.05 | 56.74 | 75.99 | 88.71 | 92.46 | |
| GOALBackbone=ViT-L/142026.05 | 54.67 | 74.86 | 87.12 | 90.94 | |
| GOALBackbone=ViT-B/16, Zero-shot=true2026.05 | 51.87 | 71.1 | 84.78 | 89.12 | |
| Long-CLIPBackbone=ViT-L/142026.05 | 46.53 | 72.64 | 86.48 | 90.38 | |
| EVA-CLIPBackbone=ViT-L/142026.05 | 44.7 | 64.75 | 79.94 | 85.41 | |
| Long-CLIPBackbone=ViT-B/16, Zero-shot=true2026.05 | 42.21 | 67.21 | 83.21 | 88.21 | |
| EVA-CLIPBackbone=ViT-B/16, Zero-shot=true2026.05 | 40.71 | 61.14 | 77.73 | 83.6 | |
| FineCLIPBackbone=ViT-B/16, Zero-shot=true2026.05 | 36.48 | 58.2 | 76.14 | 82.74 | |
| FineCLIPBackbone=ViT-L/142026.05 | 35.54 | 57.04 | 75.09 | 81.49 | |
| CLIPBackbone=ViT-B/16, Zero-shot=true2026.05 | 35.07 | 54.28 | 71.84 | 78.91 | |
| CLIPBackbone=ViT-L/142026.05 | 33.37 | 51.51 | 68.25 | 75.7 |