A Text-to-Text Model for Multilingual Offensive Language Identification

TL;DR

This paper introduces a novel encoder-decoder T5 model for multilingual offensive language detection, outperforming existing models on various benchmarks and enabling effective multilingual offensive content identification.

Contribution

It presents the first encoder-decoder T5-based model for offensive language detection, trained on large datasets, and demonstrates superior performance over existing models in multiple languages.

Findings

T5 model outperforms BERT and XLNet in offensive language detection.

Multilingual mT5 achieves state-of-the-art results across six languages.

The models are publicly available for community use.

Abstract

The ubiquity of offensive content on social media is a growing cause for concern among companies and government organizations. Recently, transformer-based models such as BERT, XLNET, and XLM-R have achieved state-of-the-art performance in detecting various forms of offensive content (e.g. hate speech, cyberbullying, and cyberaggression). However, the majority of these models are limited in their capabilities due to their encoder-only architecture, which restricts the number and types of labels in downstream tasks. Addressing these limitations, this study presents the first pre-trained model with encoder-decoder architecture for offensive language identification with text-to-text transformers (T5) trained on two large offensive language identification datasets; SOLID and CCTK. We investigate the effectiveness of combining two datasets and selecting an optimal threshold in semi-supervised instances in SOLID in the T5 retraining step. Our pre-trained T5 model outperforms other transformer-based models fine-tuned for offensive language detection, such as fBERT and HateBERT, in multiple English benchmarks. Following a similar approach, we also train the first multilingual pre-trained model for offensive language identification using mT5 and evaluate its performance on a set of six different languages (German, Hindi, Korean, Marathi, Sinhala, and Spanish). The results demonstrate that this multilingual model achieves a new state-of-the-art on all the above datasets, showing its usefulness in multilingual scenarios. Our proposed T5-based models will be made freely available to the community.