CIFAR
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
CIFAR100
99.83ASR
38
CIFAR-10 with Places365 (semantic OOD) and CIFAR-10-C (covariate OOD) (test)
88.93OOD Accuracy
38
CIFAR-10 ID Textures semantic OOD CIFAR-10-C covariate OOD
90.51OOD Accuracy
38
CIFAR-100
65Top-1 Accuracy
38
CIFAR-10
65.88AA Accuracy
38
CIFAR-10-C (test)
2.1ECE
37
DVS-CIFAR10
86.7Accuracy
37
CIFAR100
67.9Accuracy
37
CIFAR-100 Dir-0.5
73.02Accuracy
37
CIFAR-10 (IID)
0.952Average BA
37
CIFAR10
94.3Clean Accuracy
37
CIFAR-100 50% symmetric label noise
79.2Final Accuracy
37
CIFAR-10 sequential (test)
91.13Accuracy
37
CIFAR-10-C v1 (test)
8.954Classification Error
37
Cifar-100
78FDR
36
Cifar-10
97FDR
36
CIFAR-10
6.76AUC
36
CIFAR-10 8000 images (test)
93.2Accuracy
36
CIFAR-10-LTN
95.7Accuracy
36
CIFAR-10
88.27Performance at 60% Threshold
36
CIFAR10
0.04Attack Cost
36
CIFAR10 3 Tasks
74.21Accuracy
36
CIFAR-10 (test)
98.12Accuracy
36
CIFAR-100
85.87Accuracy
36
CIFAR10
59.5Accuracy (%)
36