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

Self-evaluation research benchmarks · SOTA2 Research

Self-evaluation

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

0.747AUROC

36

Feb 26, 2026

80.2AUROC

36

Feb 26, 2026

0.886AUROC

36

Feb 26, 2026

77AUROC

36

Feb 26, 2026

74.1Top-5 Accuracy

3

Jun 4, 2026

Arena-Hard v2.0

74.8Top-5 Accuracy

3

Jun 4, 2026

HelpSteer2 (val)

87.8Top-5 Accuracy

3

Jun 4, 2026

Popular datasets

Arena-Hard MMVet CVBench Helpsteer2 ViLP

Follow for updates

Join SOTA2 Contact