Policy Violation Detection
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
86F1 Score
12
Feb 26, 2026
58True Positives (TP)
5
Apr 15, 2026
0.033Violation Rate
5
Apr 15, 2026
54Violated Count
5
Apr 15, 2026
69True Positives
4
Jun 26, 2026
234True Positives
3
Jun 26, 2026