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

HexPhi

Benchmarks

Task NameDataset NameSOTA ResultTrend
Safety EvaluationHexPhi
Harmfulness2
140
Harmful Content Safety EvaluationHexPHI
Attack Success Rate (ASR)1.67
20
Showing 2 of 2 rows