List your company
ResearchFormerly WizWand

Image Classification on KMNIST standard (test)

89.68AccuracyOurs
68.453673.964379.47584.9857May 8, 2026
Updated 5mo ago

Evaluation results

MethodLinks
OursP(s=1)=0.5, Base loss=MAE, Backbone=CNN2026.0589.68
Trans-PRODENP(s=1)=0.5, Base loss=CCE, Backbone=CNN2026.0589.53
OursP(s=1)=0.7, Base loss=MAE, Backbone=CNN2026.0589.26
Trans-PRODENP(s=1)=0.7, Base loss=CCE, Backbone=CNN2026.0589.22
OursP(s=1)=0.3, Base loss=MAE, Backbone=CNN2026.0589.13
Trans-PRODENP(s=1)=0.3, Base loss=CCE, Backbone=CNN2026.0589.01
Trans-MCL (MAE)P(s=1)=0.5, Base loss=MAE, Backbone=CNN2026.0588.33
Trans-MCL (MAE)P(s=1)=0.7, Base loss=MAE, Backbone=CNN2026.0586.73
Trans-MCL (MAE)P(s=1)=0.3, Base loss=MAE, Backbone=CNN2026.0584.64
Trans-MCL (EXP)P(s=1)=0.5, Base loss=EXP, Backbone=CNN2026.0584.34
Trans-MCL (EXP)P(s=1)=0.7, Base loss=EXP, Backbone=CNN2026.0582.35
Trans-MCL (EXP)P(s=1)=0.3, Base loss=EXP, Backbone=CNN2026.0581.79
Trans-PLL-AvgP(s=1)=0.5, Base loss=MAE, Backbone=CNN2026.0573.1
Trans-PLL-AvgP(s=1)=0.3, Base loss=MAE, Backbone=CNN2026.0569.63
Trans-PLL-AvgP(s=1)=0.7, Base loss=MAE, Backbone=CNN2026.0569.27