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

MalwareBench

Benchmarks

Task NameDataset NameSOTA ResultTrend
Jailbreak AttackMalwareBench
Attack Success Rate (ASR)81.35
82
Malicious Code GenerationMalwareBench
ASR91.46
40
Safety EvaluationMalwareBench
Attack Success Rate (ASR)3.02
18
Malicious Package DetectionMalwareBench adv (test)
Recall (TPR) @ 1% FPR86.81
5
Malicious Package DetectionMalwareBench (test)
Recall @ 1% FPR95.64
5
Showing 5 of 5 rows