List your company

Detail Perception on MMVP VLM

26.7Orientation and Direction AccuracyDIVA
-1.0686.14113.3520.559Mar 5, 2026

Evaluation results

MethodLinks
DIVACLIP Backbone=OpenAI ViT-L-14, Resol.=336, #Params=427.9M2026.0326.72033.313.313.346.726.76.74025.2
OriginalCLIP Backbone=SigLIP ViT-SO-14, Resol.=224, #Params=877.4M2026.0326.72053.3402066.7402053.337.8
DIVACLIP Backbone=SigLIP ViT-SO-14, Resol.=384, #Params=878.0M2026.0326.733.353.326.713.3804026.746.738.5
GenHancerCLIP Backbone=SigLIP ViT-SO-14, Resol.=384, #Params=878.0M2026.0326.72066.733.313.386.74026.746.740
OursCLIP Backbone=SigLIP ViT-SO-14, Resol.=384, #Params=878.0M2026.0326.733.36033.326.7804033.346.742.2
OursCLIP Backbone=MetaCLIP ViT-L-14, Resol.=224, #Params=427.6M2026.03202053.313.326.78033.313.333.332.6
GenHancerCLIP Backbone=MetaCLIP ViT-H-14, Resol.=224, #Params=986.1M2026.03202066.726.726.766.733.32053.337
OursCLIP Backbone=MetaCLIP ViT-H-14, Resol.=224, #Params=986.1M2026.03202066.726.726.766.733.326.753.337.8
GenHancerCLIP Backbone=SigLIP ViT-SO-14, Resol.=224, #Params=877.4M2026.03202066.7602086.7401353.342.2
OriginalCLIP Backbone=SigLIP ViT-SO-14, Resol.=384, #Params=878.0M2026.032026.76033.313.366.733.326.753.337
un2CLIPCLIP Backbone=SigLIP ViT-SO-14, Resol.=384, #Params=878.0M2026.0320206046.726.773.34026.76041.5
OriginalCLIP Backbone=OpenAI ViT-L-14, Resol.=224, #Params=427.6M2026.0313.313.3202013.353.3206.713.319.2
DIVACLIP Backbone=OpenAI ViT-L-14, Resol.=224, #Params=427.6M2026.0313.320406.72053.346.72013.325.9
GenHancerCLIP Backbone=OpenAI ViT-L-14, Resol.=224, #Params=427.6M2026.0313.333.333.3206.773.346.7204031.8
OursCLIP Backbone=OpenAI ViT-L-14, Resol.=224, #Params=427.6M2026.0313.333.333.326.76.773.346.72046.733.3
OursCLIP Backbone=OpenAI ViT-L-14, Resol.=336, #Params=427.9M2026.0313.326.733.3206.773.346.733.326.731.1
OriginalCLIP Backbone=MetaCLIP ViT-L-14, Resol.=224, #Params=427.6M2026.0313.36.766.76.733.346.7206.713.323.7
GenHancerCLIP Backbone=MetaCLIP ViT-L-14, Resol.=224, #Params=427.6M2026.0313.32053.313.326.78033.313.333.331.8
DIVACLIP Backbone=MetaCLIP ViT-H-14, Resol.=224, #Params=986.1M2026.0313.32053.333.313.366.733.313.34031.8
DIVACLIP Backbone=SigLIP ViT-SO-14, Resol.=224, #Params=877.4M2026.0313.326.76046.713.373.353.326.753.340.7
OursCLIP Backbone=SigLIP ViT-SO-14, Resol.=224, #Params=877.4M2026.0313.326.773.3602080402053.343
GenHancerCLIP Backbone=OpenAI ViT-L-14, Resol.=336, #Params=427.9M2026.036.72033.3206.773.353.326.726.729.6
un2CLIPCLIP Backbone=OpenAI ViT-L-14, Resol.=336, #Params=427.9M2026.036.733.346.713.313.38040202030.4
DIVACLIP Backbone=MetaCLIP ViT-L-14, Resol.=224, #Params=427.6M2026.036.76.760026.766.720204027.4
OriginalCLIP Backbone=MetaCLIP ViT-H-14, Resol.=224, #Params=986.1M2026.036.713.36013.36.753.326.713.333.325.2
un2CLIPCLIP Backbone=OpenAI ViT-L-14, Resol.=224, #Params=427.6M2026.03033.346.726.713.380402033.332.6
OriginalCLIP Backbone=OpenAI ViT-L-14, Resol.=336, #Params=427.9M2026.0302040206.72033.36.733.320