Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Malicious Prompt Detection on AdvBench (subset)
Loading...
30
Fires Rate
v2 rubric (deterministic synthesis function)
28.5
29.25
30
30.75
Jun 3, 2026
Fires Rate
Held-out TP Rate
Updated 1mo ago
Evaluation Results
Method
Method
Links
Fires Rate
Held-out TP Rate
v2 rubric (deterministic synthesis function)
Prompts=50, Rules eval=3
2026.06
30
60
Feedback
Search any
task
Search any
task
