Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Embedding Space Attacks on ES attacks
Loading...
89.8
ASR
GD
6.392
28.046
49.7
71.354
Apr 12, 2026
ASR
Benign Refusal Rate
MMLU Accuracy Delta
Updated 5d ago
Evaluation Results
Method
Method
Links
ASR
Benign Refusal Rate
MMLU Accuracy Delta
GD
2026.04
89.8
6.2
-0.7
Gemma IT
2026.04
86.9
0
0
CB
2026.04
52.3
3.1
-3.6
LIRA*
2026.04
18.1
2.3
-2.3
AdLIRA*
2026.04
9.6
3.1
-5.9
Feedback
Search any
task
Search any
task
