Response Classification
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
96.7Financial Score
40
Mar 4, 2026
84.8F1 Score
39
Mar 4, 2026
86.3F1 Score
32
Mar 4, 2026
89.9BeaverT Score
16
Mar 4, 2026
98.43F1 Score
16
Feb 27, 2026
93.17F1 Score
16
Feb 27, 2026
Generic Response Classification Suite (Aegis2.0, Beavertails, SEval, SafeRLHF, Think, WildG, XSTest)
86.5Aegis2.0
16
Feb 26, 2026
75.4AUPRC (English)
16
Feb 26, 2026
60.5Pos Score
13
Apr 24, 2026
55.3Pos
13
Apr 24, 2026
97.7AUPRC
9
Feb 26, 2026
89.7AUPRC
9
Feb 26, 2026
92.1AUPRC
9
Feb 26, 2026
94.1F1 Score
5
Jun 1, 2026