Zero Shot Transfer Image Classification On 3
Métriques
Accuracy (Private)
Accuracy (Public)
Résultats
Résultats de performance de divers modèles sur ce benchmark
Tableau comparatif
| Nom du modèle | Accuracy (Private) | Accuracy (Public) |
|---|---|---|
| lit-zero-shot-transfer-with-locked-image-text | 78.7 | 66.6 |
| internvl-scaling-up-vision-foundation-models | 77.3 | - |
| pali-a-jointly-scaled-multilingual-language | 64.46 | - |
| coca-contrastive-captioners-are-image-text | 80.7 | - |
| altclip-altering-the-language-encoder-in-clip | 68.1 | - |
| pali-a-jointly-scaled-multilingual-language | 80.6 | - |
| scaling-up-visual-and-vision-language | 70.1 | - |
| eva-clip-improved-training-techniques-for | 75.7 | - |
| Modèle 9 | 81.2 | - |
| eva-clip-18b-scaling-clip-to-18-billion | 77.9 | - |
| scaling-vision-transformers-to-22-billion | 80.9 | - |
| combined-scaling-for-zero-shot-transfer | 80.6 | - |
| learning-transferable-visual-models-from | 70.1 | - |