@inproceedings{bose-etal-2021-generalisability,
    title = "Generalisability of Topic Models in Cross-corpora Abusive Language Detection",
    author = "Bose, Tulika  and
      Illina, Irina  and
      Fohr, Dominique",
    editor = "Feldman, Anna  and
      Da San Martino, Giovanni  and
      Leberknight, Chris  and
      Nakov, Preslav",
    booktitle = "Proceedings of the Fourth Workshop on NLP for Internet Freedom: Censorship, Disinformation, and Propaganda",
    month = jun,
    year = "2021",
    address = "Online",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-emnlp/2021.nlp4if-1.8/",
    doi = "10.18653/v1/2021.nlp4if-1.8",
    pages = "51--56",
    abstract = "Rapidly changing social media content calls for robust and generalisable abuse detection models. However, the state-of-the-art supervised models display degraded performance when they are evaluated on abusive comments that differ from the training corpus. We investigate if the performance of supervised models for cross-corpora abuse detection can be improved by incorporating additional information from topic models, as the latter can infer the latent topic mixtures from unseen samples. In particular, we combine topical information with representations from a model tuned for classifying abusive comments. Our performance analysis reveals that topic models are able to capture abuse-related topics that can transfer across corpora, and result in improved generalisability."
}