Image-to-Text Retrieval on COCO English 5K (test)
74.9R@1InternVL-G
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| InternVL-Gmulti-lingual=true, Zero-shot=true2023.12 | 74.9 | 91.3 | 95.2 | — | |
| InternVL-Cmulti-lingual=true, Zero-shot=true2023.12 | 70.6 | 89 | 93.5 | — | |
| EVA-02-CLIP-E+multi-lingual=false, Zero-shot=true2023.12 | 68.8 | 87.8 | 92.8 | — | |
| EVA-01-CLIP-g+multi-lingual=false, Zero-shot=true2023.12 | 68.2 | 87.5 | 92.5 | — | |
| OpenCLIP-Gmulti-lingual=false, Zero-shot=true2023.12 | 67.3 | 86.9 | 92.6 | — | |
| OpenCLIP-gmulti-lingual=false, Zero-shot=true2023.12 | 66.4 | 86 | 91.8 | — | |
| CoCamulti-lingual=false, Zero-shot=true2023.12 | 66.3 | 86.2 | 91.8 | — | |
| OpenCLIP-Hmulti-lingual=false, Zero-shot=true2023.12 | 66 | 86.1 | 91.9 | — | |
| OpenCLIP-XLM-R-Hmulti-lingual=true, Zero-shot=true2023.12 | 65.9 | 86.2 | 92.2 | — | |
| Florencemulti-lingual=false, Zero-shot=true2023.12 | 64.7 | 85.9 | — | — | |
| ONE-PEACEmulti-lingual=false, Zero-shot=true2023.12 | 64.7 | 86 | 91.9 | — |