Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

REALTOX

Benchmarks

Task NameDataset NameSOTA ResultTrend
Toxicity DetectionRealTox Challenging
Accuracy98.92
12
Toxicity DetectionRealTox Standard
Accuracy87.56
12
Toxicity ControlREALTOX
Accuracy0.925
11
Generation QualityRealTox
PPL33.97
6
Showing 4 of 4 rows