@inproceedings{kotyushev-etal-2021-mipt,
    title = "{MIPT}-{NSU}-{UTMN} at {S}em{E}val-2021 Task 5: Ensembling Learning with Pre-trained Language Models for Toxic Spans Detection",
    author = "Kotyushev, Mikhail  and
      Glazkova, Anna  and
      Morozov, Dmitry",
    editor = "Palmer, Alexis  and
      Schneider, Nathan  and
      Schluter, Natalie  and
      Emerson, Guy  and
      Herbelot, Aurelie  and
      Zhu, Xiaodan",
    booktitle = "Proceedings of the 15th International Workshop on Semantic Evaluation (SemEval-2021)",
    month = aug,
    year = "2021",
    address = "Online",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/jlcl-multiple-ingestion/2021.semeval-1.124/",
    doi = "10.18653/v1/2021.semeval-1.124",
    pages = "913--918",
    abstract = "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."
}