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

HoliSafe-Bench

Benchmarks

Task NameDataset NameSOTA ResultTrend
Harmful Content DetectionHoliSafe-Bench
AUPRC75.6
49
Safety ClassificationHoliSafe-Bench
AUROC0.783
49
Safety ClassificationHoliSafe-Bench
ECE8.4
21
Showing 3 of 3 rows