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

Classification research benchmarks – Page 64 · SOTA2 Research

Classification

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

0.164Empirical Risk (Rs(h))

11

Feb 26, 2026

95Balanced Accuracy

11

Feb 26, 2026

78BAcc

11

Feb 26, 2026

CLIP 21-dataset Zero-Shot standard (test val)

45ImageNet Accuracy

11

Feb 27, 2026

32.86Error Rate

11

Feb 26, 2026

ScanObjectNN OBJ_ONLY original objects

PointMLP w/ PointCMT

93.2OA

11

Feb 26, 2026

CIFAR-10 50% seen 50% novel

94.9Seen Accuracy

11

Feb 26, 2026

Omniglot (test)

99.84Top-1 Acc

11

Feb 26, 2026

95.509Balanced Accuracy

11

Jul 7, 2026

98.54Balanced Accuracy

11

Jul 7, 2026

99.62Balanced Acc

11

Jul 7, 2026

88.25Balanced Accuracy

11

Jul 7, 2026

90Balanced Accuracy

11

Jul 7, 2026

SubTab (+Noise)

71.13Accuracy

11

Feb 26, 2026

SubTab (+Distance loss)

84.64Accuracy

11

Feb 26, 2026

82.57Accuracy

11

Feb 26, 2026

99.53Accuracy

11

Feb 26, 2026

94.23Accuracy

11

Feb 26, 2026

Clothing1M 10k clean (test)

74.02Accuracy

11

Feb 26, 2026

Reuters 36 classes (test)

70bACC

11

Feb 26, 2026

Twitter Natural language (test)

78.8bACC

11

Feb 26, 2026

SUN397 Natural imbalance (test)

33.3bACC

11

Feb 26, 2026

CelebA 5-way hair color (test)

75.9bACC

11

Feb 26, 2026

90.2Accuracy

11

Feb 26, 2026

Semantic Genesis 3D

98.47AUC

11

Feb 26, 2026

Page 64 of 202

Popular datasets

CIFAR MNIST Adult ImageNet Wine

Follow for updates

Join SOTA2 Contact