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

Prompt Injection Attack on PaperGuard

80Attack Success Rate (ASR)

Claude-sonnet-4.5

44.6453.826372.18Jun 10, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.06
8014.141.13-2.62
2026.06
769.950-3.26
2026.06
747.930.27-1.88
2026.06
737.110-2.43
2026.06
726.650.13-1.37
2026.06
689.40.43-2.27
2026.06
644.160.3-1.39
2026.06
465.050.460.11