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

MMDS

Benchmarks

Task NameDataset NameSOTA ResultTrend
Content ModerationMMDS (test)
Accuracy95.76
27
Multimodal SafetyMMDS-R
F1 Score96.53
15
Multimodal SafetyMMDS-Q
F1 Score99.13
15
Attack Success Rate EvaluationMMDS MMRT (test)
ASR100
7
Showing 4 of 4 rows