Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Safety Alignment on RAB Violence (adversarial prompts)
Loading...
1
ASR
AEGIS
-0.92
12.04
25
37.96
Jul 7, 2026
ASR
Updated 17d ago
Evaluation Results
Method
Method
Links
ASR
AEGIS
Category=AEGIS (Ours),...
2026.07
1
AEGIS
Category=AEGIS (Ours),...
2026.07
3
SafeClip
Category=Input-Space
2026.07
4
SLD-Strong
Category=Trigger-Specific
2026.07
7
SAFREE
Category=Input-Space
2026.07
8
CAD
Category=Feature Pruning
2026.07
8
SafeGuider
Category=Input-Space
2026.07
10
UCE
Category=Trigger-Specific
2026.07
11
SDID
Category=Trigger-Specific
2026.07
11
Receler
Category=Trigger-Specific
2026.07
15
SD 2.1
Category=No Alignment
2026.07
22
SLD-Medium
Category=Trigger-Specific
2026.07
23
ESD-u
Category=Trigger-Specific
2026.07
24
Diff-QuickFix
Category=Trigger-Specific
2026.07
25
SLD-Weak
Category=Trigger-Specific
2026.07
39
SD 1.4
Category=No Alignment
2026.07
49
Feedback
Search any
task
Search any
task