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

XGuard

Benchmarks

Task NameDataset NameSOTA ResultTrend
Text Query SafetyXGuard (test)
F1 Score92.66
28
Jailbreak DetectionXGuard (train)
AUC0.908
9
Showing 2 of 2 rows