MIPT-NSU-UTMN at SemEval-2021 Task 5: Ensembling Learning with Pre-trained Language Models for Toxic Spans Detection
This work addresses toxic content detection for online moderation, but it is incremental as it builds on existing pre-trained models and ensemble techniques.
The paper tackled the problem of detecting toxic spans in text by developing ensemble models using BERT-based architectures and post-processing, achieving an F1-score of 67.55% on test data.
This paper describes our system for SemEval-2021 Task 5 on Toxic Spans Detection. We developed ensemble models using BERT-based neural architectures and post-processing to combine tokens into spans. We evaluated several pre-trained language models using various ensemble techniques for toxic span identification and achieved sizable improvements over our baseline fine-tuned BERT models. Finally, our system obtained a F1-score of 67.55% on test data.