List your company
ResearchFormerly WizWand

Out-of-Distribution Detection on CIFAR-100 (ID) vs Places365 (Far-OOD) (OpenOOD v1.5 Test)

44.07FPR@95GradNorm
42.16455.029567.89580.7605Apr 29, 2026
Updated 6mo ago

Evaluation results

MethodLinks
GradNormBackbone=ViT-B/16, Pre-trained Model=google/vit-base-patch16-224-in21k, ID Accuracy=84.78%, Evaluation Protocol=fine-tuned2026.0444.0788.41
ViMBackbone=ViT-B/16, Pre-trained Model=google/vit-base-patch16-224-in21k, ID Accuracy=84.78%, Evaluation Protocol=fine-tuned2026.0446.5186.53
SHEBackbone=ViT-B/16, Pre-trained Model=google/vit-base-patch16-224-in21k, ID Accuracy=84.78%, Evaluation Protocol=fine-tuned2026.0446.9687.4
ReActBackbone=ViT-B/16, Pre-trained Model=google/vit-base-patch16-224-in21k, ID Accuracy=84.78%, Evaluation Protocol=fine-tuned2026.0447.4287.14
RMDSBackbone=ViT-B/16, Pre-trained Model=google/vit-base-patch16-224-in21k, ID Accuracy=84.78%, Evaluation Protocol=fine-tuned2026.0449.7984.64
KNNBackbone=ViT-B/16, Pre-trained Model=google/vit-base-patch16-224-in21k, ID Accuracy=84.78%, Evaluation Protocol=fine-tuned2026.0450.6685.39
EPDBackbone=ViT-B/16, Pre-trained Model=google/vit-base-patch16-224-in21k, ID Accuracy=84.78%, Evaluation Protocol=fine-tuned2026.0452.4485.21
MDSBackbone=ViT-B/16, Pre-trained Model=google/vit-base-patch16-224-in21k, ID Accuracy=84.78%, Evaluation Protocol=fine-tuned2026.0456.4782.55
EBOBackbone=ViT-B/16, Pre-trained Model=google/vit-base-patch16-224-in21k, ID Accuracy=84.78%, Evaluation Protocol=fine-tuned2026.0464.0783.81
MLSBackbone=ViT-B/16, Pre-trained Model=google/vit-base-patch16-224-in21k, ID Accuracy=84.78%, Evaluation Protocol=fine-tuned2026.0464.0783.58
TempScaleBackbone=ViT-B/16, Pre-trained Model=google/vit-base-patch16-224-in21k, ID Accuracy=84.78%, Evaluation Protocol=fine-tuned2026.0464.1982.13
GENBackbone=ViT-B/16, Pre-trained Model=google/vit-base-patch16-224-in21k, ID Accuracy=84.78%, Evaluation Protocol=fine-tuned2026.0464.4383.8
MSPBackbone=ViT-B/16, Pre-trained Model=google/vit-base-patch16-224-in21k, ID Accuracy=84.78%, Evaluation Protocol=fine-tuned2026.0464.5680.8
DICEBackbone=ViT-B/16, Pre-trained Model=google/vit-base-patch16-224-in21k, ID Accuracy=84.78%, Evaluation Protocol=fine-tuned2026.0465.1182.65
OpenMaxBackbone=ViT-B/16, Pre-trained Model=google/vit-base-patch16-224-in21k, ID Accuracy=84.78%, Evaluation Protocol=fine-tuned2026.0466.281.15
ASHBackbone=ViT-B/16, Pre-trained Model=google/vit-base-patch16-224-in21k, ID Accuracy=84.78%, Evaluation Protocol=fine-tuned2026.0491.7252.57