Categories Companies Research

Loading the SOTA2 catalog…

Explore

Categories Companies Research

Contribute

Join SOTA2 Contact

SOTA2

© 2026 SOTA2

Build in Bellevue, USA

WizWand is now SOTA2 Research

Papers Benchmarks Datasets Tasks

Image Classification research benchmarks – Page 221 · SOTA2 Research

Image Classification

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

CIFAR-100 beta=0.3

81.65Accuracy

7

Feb 26, 2026

ImageNet (test)

76.23Standard Accuracy

7

Feb 26, 2026

87Accuracy

7

Feb 26, 2026

61.1Accuracy

7

Feb 26, 2026

74.9Accuracy

7

Feb 26, 2026

60.7Accuracy

7

Feb 26, 2026

68.23Last Accuracy

7

Feb 26, 2026

ImageYellow 128x128 (test)

84.4Accuracy

7

Feb 26, 2026

78.07Base Accuracy

7

Feb 26, 2026

ImageNet-1K (val)

Full ImageNet-1K

69.8Accuracy

7

Feb 26, 2026

CIFAR-10 2010 (test)

62.8Accuracy

7

Feb 26, 2026

-0.037Log-likelihood

7

Feb 26, 2026

CIFAR-10 Unseen Adversaries (test)

26.1Union

7

Feb 26, 2026

30-Task Vision Merging Average

93.02Avg Accuracy

7

Feb 26, 2026

99.63Accuracy

7

Feb 26, 2026

Garbage Classification

93.36Accuracy

7

Feb 26, 2026

85.53Accuracy

7

Feb 26, 2026

98.48Accuracy

7

Feb 26, 2026

Landscape Recognition

86.83Accuracy

7

Feb 26, 2026

95.27Accuracy

7

Feb 26, 2026

94.31Accuracy

7

Feb 26, 2026

100Accuracy

7

Feb 26, 2026

ImageNet 10% labels 1k (train val)

76Top-1 Accuracy

7

Feb 26, 2026

ImageNet 1% labels 1k (train val)

66.8Top-1 Accuracy

7

Feb 26, 2026

CIFAR-10 Instance-Dependent Noise 50% (test)

76.87Accuracy

7

Feb 26, 2026

Page 221 of 333

Popular datasets

CIFAR ImageNet MNIST CIFAR-10 CIFAR-100

Follow for updates

Join SOTA2 Contact