@inproceedings{dai-etal-2020-cost,
    title = "Cost-effective Selection of Pretraining Data: A Case Study of Pretraining {BERT} on Social Media",
    author = "Dai, Xiang  and
      Karimi, Sarvnaz  and
      Hachey, Ben  and
      Paris, Cecile",
    editor = "Cohn, Trevor  and
      He, Yulan  and
      Liu, Yang",
    booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2020",
    month = nov,
    year = "2020",
    address = "Online",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-emnlp/2020.findings-emnlp.151/",
    doi = "10.18653/v1/2020.findings-emnlp.151",
    pages = "1675--1681",
    abstract = "Recent studies on domain-specific BERT models show that effectiveness on downstream tasks can be improved when models are pretrained on in-domain data. Often, the pretraining data used in these models are selected based on their subject matter, e.g., biology or computer science. Given the range of applications using social media text, and its unique language variety, we pretrain two models on tweets and forum text respectively, and empirically demonstrate the effectiveness of these two resources. In addition, we investigate how similarity measures can be used to nominate in-domain pretraining data. We publicly release our pretrained models at \url{https://bit.ly/35RpTf0}."
}