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

Content Safety Detection research benchmarks · SOTA2 Research

Content Safety Detection

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

HUS-II HPAA-applied

0Hate Detection Rate

10

Jun 9, 2026

100Hate Detection Rate

10

Jun 9, 2026

prompt–response verification framework

96.6F1 Score

4

Jun 26, 2026

prompt–response verification framework

96.4F1 Score

2

Jun 26, 2026

prompt–response verification framework

95.9F1 Score

2

Jun 26, 2026

Aegis-Safety 2.0

prompt–response verification framework

95.1F1 Score

2

Jun 26, 2026

prompt–response verification framework

96.2F1 Score

2

Jun 26, 2026

prompt–response verification framework

95.6F1 Score

2

Jun 26, 2026

HarmBench cyber

prompt–response verification framework

92.8F1 Score

2

Jun 26, 2026

prompt–response verification framework

92.2F1 Score

2

Jun 26, 2026

prompt–response verification framework

93.1F1 Score

2

Jun 26, 2026

prompt–response verification framework

98.1F1 Score

2

Jun 26, 2026

prompt–response verification framework

92.8F1 Score

2

Jun 26, 2026

deepset prompt-injections

prompt–response verification framework

93.9F1 Score

2

Jun 26, 2026

prompt–response verification framework

93.1F1 Score

2

Jun 26, 2026

prompt–response verification framework

95.8F1 Score

2

Jun 26, 2026

prompt–response verification framework

96.2F1 Score

2

Jun 26, 2026

Popular datasets

HarmBench WildGuardMix JailbreakBench Wildjailbreak BeaverTails

Follow for updates

Join SOTA2 Contact