Meta AI at Arabic Hate Speech 2022: MultiTask Learning with Self-Correction for Hate Speech Classification
This work addresses hate speech classification for Arabic social media, but it is incremental as it builds on existing methods with specific enhancements.
The paper tackled Arabic fine-grained hate speech detection by developing an ensemble model with multitask learning and self-consistency correction, achieving 82.7% accuracy on the hate speech subtask, a 3.4% relative improvement over previous work.
In this paper, we tackle the Arabic Fine-Grained Hate Speech Detection shared task and demonstrate significant improvements over reported baselines for its three subtasks. The tasks are to predict if a tweet contains (1) Offensive language; and whether it is considered (2) Hate Speech or not and if so, then predict the (3) Fine-Grained Hate Speech label from one of six categories. Our final solution is an ensemble of models that employs multitask learning and a self-consistency correction method yielding 82.7% on the hate speech subtask -- reflecting a 3.4% relative improvement compared to previous work.