List your company
ResearchFormerly WizWand

Image Classification on CIFAR-10 Bucket 1 Retain

98.46AccuracyRetain-only Retain
92.719294.209695.797.1904Apr 9, 2026
Updated 7mo ago

Evaluation results

MethodLinks
Retain-only RetainEvaluation Head=Output, Backbone=ViT-S/16, Pre-training=ImageNet2026.0498.46
Retain-only RetainEvaluation Head=Linear Probe, Backbone=ViT-S/16, Pre-training=ImageNet2026.0498.28
OriginalEvaluation Head=Output, Backbone=ViT-S/16, Pre-training=ImageNet2026.0498.2
OriginalEvaluation Head=Linear Probe, Backbone=ViT-S/16, Pre-training=ImageNet2026.0498.17
Retain-only RetainEvaluation Head=NCC, Backbone=ViT-S/16, Pre-training=ImageNet2026.0498.12
OriginalEvaluation Head=NCC, Backbone=ViT-S/16, Pre-training=ImageNet2026.0498.06
Random-label with CMFEvaluation Head=Output, Backbone=ViT-S/16, Pre-training=ImageNet2026.0497.38
Random-label with CMFEvaluation Head=NCC, Backbone=ViT-S/16, Pre-training=ImageNet2026.0497.35
Salun with CMFEvaluation Head=Output, Backbone=ViT-S/16, Pre-training=ImageNet2026.0497.3
Salun with CMFEvaluation Head=NCC, Backbone=ViT-S/16, Pre-training=ImageNet2026.0497.26
Random-label with CMFEvaluation Head=Linear Probe, Backbone=ViT-S/16, Pre-training=ImageNet2026.0497.2
Salun with CMFEvaluation Head=Linear Probe, Backbone=ViT-S/16, Pre-training=ImageNet2026.0497.11
NegGrad+ with CMFEvaluation Head=Linear Probe, Backbone=ViT-S/16, Pre-training=ImageNet2026.0493.76
NegGrad+ with CMFEvaluation Head=NCC, Backbone=ViT-S/16, Pre-training=ImageNet2026.0493.04
NegGrad+ with CMFEvaluation Head=Output, Backbone=ViT-S/16, Pre-training=ImageNet2026.0492.94