Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Attack Detection on HotpotQA and FEVER
Loading...
0.94
AUROC
RSP-M
0.7112
0.7706
0.83
0.8894
Dec 16, 2025
AUROC
FPR (@95% TPR)
Updated 4d ago
Evaluation Results
Method
Method
Links
AUROC
FPR (@95% TPR)
RSP-M
2025.12
0.94
6.2
LLM-Judge
backbone=gemini-2.5
2025.12
0.76
31.5
Step Count (Simple)
Threshold=N > 10
2025.12
0.72
42
Feedback
Search any
task
Search any
task
