Why Is It Hate Speech? Masked Rationale Prediction for Explainable Hate Speech Detection

Benchmark Model Rank Results
hate-speech-detection-on-hatexplainBERT-MRP#1AUROC: 0.862Macro F1: 0.699Accuracy: 0.704
hate-speech-detection-on-hatexplainBERT-RP#2AUROC: 0.853Macro F1: 0.693Accuracy: 0.707