Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Indirect Prompt Injection Robustness on Agent Security Bench IPI
Loading...
2
Attack Success Rate (ASR)
GPT-5
-0.6
16.95
34.5
52.05
Mar 3, 2026
Attack Success Rate (ASR)
Response Rate (RR)
Updated 1mo ago
Evaluation Results
Method
Method
Links
Attack Success Rate (ASR)
Response Rate (RR)
GPT-5
Safety Scaffolding=MOS...
2026.03
2
65
GPT-5
Safety Scaffolding=Bas...
2026.03
3
0
GPT-4o
Safety Scaffolding=MOS...
2026.03
27
63
Phi-4
Safety Scaffolding=Bas...
2026.03
28
68
Qwen2.5-7B
Safety Scaffolding=MOS...
2026.03
33
61
Phi-4
Safety Scaffolding=MOS...
2026.03
39
54
Qwen2.5-7B
Safety Scaffolding=Bas...
2026.03
40
44
Qwen3-4B-Think
Safety Scaffolding=MOS...
2026.03
43
42
Qwen3-4B-Think
Safety Scaffolding=Bas...
2026.03
46
31
GPT-4o
Safety Scaffolding=Bas...
2026.03
67
0
Feedback
Search any
task
Search any
task