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

MM-Safety Bench

Benchmarks

Task NameDataset NameSOTA ResultTrend
Jailbreak Safety EvaluationMM-Safety Bench (test)
Average ASR0.18
56
Jailbreaking Attack DefenseMM-Safety-Bench (Turn 3)
Attack Success Rate (ASR)0
10
Jailbreaking Attack DefenseMM-Safety-Bench Turn 2
ASR0.58
5
Showing 3 of 3 rows