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

Zero-shot Classification research benchmarks – Page 4 · SOTA2 Research

Zero-shot Classification

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

ETHEC distribution shift

83.8AUC

5

Feb 26, 2026

ETHEC (in-distribution)

88.8AUC

5

Feb 26, 2026

CELEBA distribution shift

69.7AUC

5

Feb 26, 2026

CELEBA (in-distribution)

85.3AUC

5

Feb 26, 2026

ILSVRC conventional setting 2012 2010 (test)

11Hit@1

5

Feb 26, 2026

Evaluation Suite Zero-shot (ARCC, ARCE, BOOLQ, HSWAG, PIQA)

70.83Average Accuracy

4

Jun 4, 2026

BREEDS-structured ImageNet-1k (test)

73.8Level 1 Score

4

Feb 26, 2026

71.87Accuracy

4

Feb 26, 2026

Flowers 102 (unseen)

81.26Accuracy

4

Feb 26, 2026

FGVC Aircrafts (unseen)

40.75Accuracy

4

Feb 26, 2026

Stanford Cars (unseen)

75.78Accuracy

4

Feb 26, 2026

DataComp medium-scale (test)

Mimic Score Top 30%

26.8Average Score

3

May 27, 2026

27.25Accuracy

3

Mar 16, 2026

WOODS domain averages

78.2CAP (EEG)

3

Mar 10, 2026

Internal clinical dataset 30 findings (test)

74.1Average F1 Score

3

Mar 5, 2026

73.9Accuracy

3

Feb 26, 2026

Winogrande v1.1 (val)

77Accuracy

3

Feb 26, 2026

48.4Average Accuracy (7-task)

2

Jun 17, 2026

Average 8 image & 4 video datasets

75.7Top-1 Accuracy

2

Feb 26, 2026

76.7Average F1 Score

1

Mar 5, 2026

External clinical dataset 30 findings (test)

64.7Average F1 Score

1

Mar 5, 2026

35.1Accuracy

1

Feb 26, 2026

Page 4 of 4

Popular datasets

ImageNet COCO CIFAR Evaluation Suite CUB

Follow for updates

Join SOTA2 Contact