WLV-RIT at SemEval-2021 Task 5: A Neural Transformer Framework for Detecting Toxic Spans

TL;DR

This paper introduces a neural transformer-based framework for detecting toxic spans in social media posts, achieving competitive results and providing an open-source multilingual tool for offensive content detection.

Contribution

It presents a novel neural transformer approach for toxic span detection and introduces MUDES, an open-source multilingual framework for offensive span identification.

Findings

Best model achieved 0.68 F1-Score on the dataset

Developed an open-source multilingual detection framework

Demonstrated effectiveness in toxic span detection

Abstract

In recent years, the widespread use of social media has led to an increase in the generation of toxic and offensive content on online platforms. In response, social media platforms have worked on developing automatic detection methods and employing human moderators to cope with this deluge of offensive content. While various state-of-the-art statistical models have been applied to detect toxic posts, there are only a few studies that focus on detecting the words or expressions that make a post offensive. This motivates the organization of the SemEval-2021 Task 5: Toxic Spans Detection competition, which has provided participants with a dataset containing toxic spans annotation in English posts. In this paper, we present the WLV-RIT entry for the SemEval-2021 Task 5. Our best performing neural transformer model achieves an $0.68$ F1-Score. Furthermore, we develop an open-source framework for multilingual detection of offensive spans, i.e., MUDES, based on neural transformers that detect toxic spans in texts.