A. Hegde

Transformer-driven Multi-task Learning for Fake and Hateful Content Detection

Two models are proposed: Hing_MTL - a Multi-task Learning (MTL) model implemented by ensembling two pre-trained Hinglish Bidirectional Encoder Representations from Transformers (Hinglish-BERT), and Ensemble_MTL - a MTL model implemented by ensembling two pre-trained models to detect fake and hate content and identify t...