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

HateSpeech

Benchmarks

Task NameDataset NameSOTA ResultTrend
ClassificationHateSpeech18
Accuracy83.44
49
Binary ClassificationHateSpeech
Accuracy92.5
6
OOD DetectionHateSpeech (test)
Length-Matched AUROC0.634
5
Text ClassificationHateSpeech (test)
Accuracy91.5
5
Learning to DeferHateSpeech
Error8.06
4
Hate Speech DetectionHateSpeech (test)
ASR79.1
4
Social Science MeasurementHatespeech
Expected Calibration Error (ECE)0.468
2
Calibration AnalysisHateSpeech
AUROC65
2
Text Anomaly DetectionHateSpeech
AUPRC0.1805
2
Showing 9 of 9 rows