Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Malicious Prompt Detection on MMBench OOD
Loading...
0.16
FPR
VLMShield
-0.2268
2.3841
4.995
7.6059
Apr 7, 2026
FPR
Updated 9d ago
Evaluation Results
Method
Method
Links
FPR
VLMShield
Target Model=LLaVA-1.5...
2026.04
0.16
VLMShield
Target Model=Qwen2.5-V...
2026.04
0.16
VLMGuard
Target Model=Qwen2.5-V...
2026.04
2
ASTRA
Target Model=LLaVA-1.5...
2026.04
2.34
CIDER
Target Model=LLaVA-1.5...
2026.04
2.54
CIDER
Target Model=Qwen2.5-V...
2026.04
2.54
VLMGuard
Target Model=LLaVA-1.5...
2026.04
3.08
ECSO
Target Model=Qwen2.5-V...
2026.04
4.93
JailGuard
Target Model=Qwen2.5-V...
2026.04
5
ASTRA
Target Model=Qwen2.5-V...
2026.04
5.36
ECSO
Target Model=LLaVA-1.5...
2026.04
7.2
JailGuard
Target Model=LLaVA-1.5...
2026.04
8.75
MirrorCheck
Target Model=LLaVA-1.5...
2026.04
9.83
MirrorCheck
Target Model=Qwen2.5-V...
2026.04
9.83
Feedback
Search any
task
Search any
task