Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Prompt Injection Detection on PaperGuard (test)
Loading...
95
Accuracy
Chunk-based
30.832
47.491
64.15
80.809
Jun 10, 2026
Accuracy
Recall
FPR
FNR
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy
Recall
FPR
FNR
Chunk-based
Method Category=Embedd...
2026.06
95
92.86
0
7.14
LLM-as-Judge
Method Category=Propri...
2026.06
66.7
100
100
0
Embedding Classifier
Method Category=Traine...
2026.06
64.5
0
12
100
BERT Classifier
Method Category=Traine...
2026.06
38.5
0
35
100
Moderation API
Method Category=Propri...
2026.06
33.3
0
0
100
Feedback
Search any
task
Search any
task