List your company
ResearchFormerly WizWand

Image Classification on SVHN standard (test)

94.5AccuracyOurs
65.556873.070980.58588.0991May 8, 2026
Updated 6mo ago

Evaluation results

MethodLinks
OursP(s=1)=0.3, Base loss=MAE, Backbone=ResNet-342026.0594.5
OursP(s=1)=0.7, Base loss=MAE, Backbone=ResNet-342026.0594.48
Trans-PRODENP(s=1)=0.7, Base loss=CCE, Backbone=ResNet-342026.0593.02
Trans-PRODENP(s=1)=0.3, Base loss=CCE, Backbone=ResNet-342026.0592.56
Trans-MCL (EXP)P(s=1)=0.3, Base loss=EXP, Backbone=ResNet-342026.0592.26
Trans-MCL (EXP)P(s=1)=0.7, Base loss=EXP, Backbone=ResNet-342026.0592.07
Trans-PRODENP(s=1)=0.5, Base loss=CCE, Backbone=ResNet-342026.0591.16
Trans-MCL (EXP)P(s=1)=0.5, Base loss=EXP, Backbone=ResNet-342026.0590.23
OursP(s=1)=0.5, Base loss=MAE, Backbone=ResNet-342026.0589.97
Trans-PLL-AvgP(s=1)=0.3, Base loss=MAE, Backbone=ResNet-342026.0580.62
Trans-PLL-AvgP(s=1)=0.5, Base loss=MAE, Backbone=ResNet-342026.0580.58
Trans-MCL (MAE)P(s=1)=0.3, Base loss=MAE, Backbone=ResNet-342026.0578.83
Trans-PLL-AvgP(s=1)=0.7, Base loss=MAE, Backbone=ResNet-342026.0578.82
Trans-MCL (MAE)P(s=1)=0.5, Base loss=MAE, Backbone=ResNet-342026.0573.3
Trans-MCL (MAE)P(s=1)=0.7, Base loss=MAE, Backbone=ResNet-342026.0566.67