Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Image-based Jailbreak on HADES OOD
Loading...
80.02
Attack Success Rate (ASR)
SelfReminder
-0.9856
20.0447
41.075
62.1053
Apr 7, 2026
Attack Success Rate (ASR)
Updated 9d ago
Evaluation Results
Method
Method
Links
Attack Success Rate (ASR)
SelfReminder
Base VLM=Qwen2.5-VL-7B...
2026.04
80.02
SelfReminder
Base Model=LLaVA-1.5-13B
2026.04
75.32
CIDER
Base VLM=Qwen2.5-VL-7B...
2026.04
51.86
CIDER
Base Model=LLaVA-1.5-13B
2026.04
51.86
JailGuard
Base Model=LLaVA-1.5-13B
2026.04
38.33
ECSO
Base Model=LLaVA-1.5-13B
2026.04
31.32
ECSO
Base VLM=Qwen2.5-VL-7B...
2026.04
30.74
MirrorCheck
Base VLM=Qwen2.5-VL-7B...
2026.04
23.09
MirrorCheck
Base Model=LLaVA-1.5-13B
2026.04
23.09
VLMGuard
Base Model=LLaVA-1.5-13B
2026.04
22.95
JailGuard
Base VLM=Qwen2.5-VL-7B...
2026.04
21.04
ASTRA
Base Model=LLaVA-1.5-13B
2026.04
14.86
VLMGuard
Base VLM=Qwen2.5-VL-7B...
2026.04
13.96
ASTRA
Base VLM=Qwen2.5-VL-7B...
2026.04
7.83
VLMShield
Base VLM=Qwen2.5-VL-7B...
2026.04
2.13
VLMShield
Base Model=LLaVA-1.5-13B
2026.04
2.13
Feedback
Search any
task
Search any
task