Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Jailbreaking Attack Defense on MM-Safety-Bench (Turn 3)
Loading...
0
Attack Success Rate (ASR)
FragGuard
-3.654
21.0105
45.675
70.3395
Jan 8, 2026
Attack Success Rate (ASR)
Attack Transfer Score (ATS)
Response Rate (RR)
Updated 1mo ago
Evaluation Results
Method
Method
Links
Attack Success Rate (ASR)
Attack Transfer Score (ATS)
Response Rate (RR)
FragGuard
Backbone=GPT-4o, Toxic...
2026.01
0
1.03
80.38
FragGuard
Backbone=Gemini-2.0-Fl...
2026.01
0.58
1.05
88.78
FragGuard
Backbone=LLaVa-7B, Tox...
2026.01
1.15
1.06
90
FragGuard
Backbone=LLaVa-13B, To...
2026.01
4.42
1.58
13.46
FragGuard
Backbone=Qwen-7B, Toxi...
2026.01
7.31
1.65
46.54
Multi-turn Jailbreaking Attack
Target Model=LLaVa-13B...
2026.01
16.54
2.17
-
Multi-turn Jailbreaking Attack
Target Model=Qwen-7B,...
2026.01
52.31
3.34
-
Multi-turn Jailbreaking Attack
Target Model=GPT-4o, T...
2026.01
67.5
3.56
-
Multi-turn Jailbreaking Attack
Target Model=Gemini-2....
2026.01
68.09
3.58
-
Multi-turn Jailbreaking Attack
Target Model=LLaVa-7B,...
2026.01
91.35
4.62
-
Feedback
Search any
task
Search any
task