Zero Shot Transfer Image Classification On 1

Accuracy (Private)

Résultats

Résultats de performance de divers modèles sur ce benchmark

Tableau comparatif

Nom du modèle	Accuracy (Private)
eva-clip-improved-training-techniques-for	82
Modèle 2	-
eva-clip-18b-scaling-clip-to-18-billion	83.8
learning-transferable-visual-models-from	76.2
pali-a-jointly-scaled-multilingual-language	85.4
learning-transferable-visual-models-from	-
learning-customized-visual-models-with	78.5
pali-a-jointly-scaled-multilingual-language	72.11
internvl-scaling-up-vision-foundation-models	83.2
learning-transferable-visual-models-from	59.6
coca-contrastive-captioners-are-image-text	86.3
Modèle 12	88.3
boldsymbol-m-2-encoder-advancing-bilingual	88.5
alternating-gradient-descent-and-mixture-of	83.9
scaling-vision-transformers-to-22-billion	85.9
Modèle 16	81.8
scaling-up-visual-and-vision-language	76.4
the-effectiveness-of-mae-pre-pretraining-for	82.1
altclip-altering-the-language-encoder-in-clip	74.5
your-diffusion-model-is-secretly-a-zero-shot	61.4
combined-scaling-for-zero-shot-transfer	85.7
the-effectiveness-of-mae-pre-pretraining-for	81.1
lit-zero-shot-transfer-with-locked-image-text	84.5