Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Safety Evaluation on XSTest (combined)

100F1 Score

ReasoningGuard

7480.7587.594.25Aug 6, 2025
Updated 27d ago

Evaluation Results

MethodLinks
2025.08
100
2025.08
97
2025.08
97
2025.08
97
2025.08
96
2025.08
95
2025.08
94
2025.08
94
2025.08
92
2025.08
92
2025.08
90
2025.08
90
2025.08
89
2025.08
89
2025.08
88
2025.08
88
2025.08
88
2025.08
87
2025.08
86
2025.08
86
2025.08
86
2025.08
85
2025.08
85
2025.08
85
2025.08
85
2025.08
84
2025.08
82
2025.08
81
2025.08
81
2025.08
80
2025.08
80
2025.08
79
2025.08
79
2025.08
75