PH
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
PH
0.007Brier Score
28
PH Toolhang (test)
86Success Rate
6
PH Square (test)
94Success Rate
6
PH Lift (test)
100Success Rate
6
PH Can (test)
98Success Rate
6
PH Seemingly Toxic
87Compliance Rate (Keyword Filter)
5