Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Attack Detection on HumanEval code (N = 100)
Loading...
1
FPR
QDM
-2.16
19.17
40.5
61.83
Jun 6, 2026
FPR
Updated 1mo ago
Evaluation Results
Method
Method
Links
FPR
QDM
Backbone Model=DS-R1-Q...
2026.06
1
RECURGUARD
Backbone Model=DS-R1-Q...
2026.06
1
Answer-Absence Proxy
Backbone Model=DS-R1-Q...
2026.06
17
Compression-Ratio Proxy
Backbone Model=DS-R1-Q...
2026.06
27
Token Budget
Backbone Model=DS-R1-Q...
2026.06
69
Output-Length Anomaly
Backbone Model=DS-R1-Q...
2026.06
80
Feedback
Search any
task
Search any
task