@inproceedings{hong-etal-2021-avocado,
    title = "{AV}oca{D}o: Strategy for Adapting Vocabulary to Downstream Domain",
    author = "Hong, Jimin  and
      Kim, TaeHee  and
      Lim, Hyesu  and
      Choo, Jaegul",
    editor = "Moens, Marie-Francine  and
      Huang, Xuanjing  and
      Specia, Lucia  and
      Yih, Scott Wen-tau",
    booktitle = "Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing",
    month = nov,
    year = "2021",
    address = "Online and Punta Cana, Dominican Republic",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/jlcl-multiple-ingestion/2021.emnlp-main.385/",
    doi = "10.18653/v1/2021.emnlp-main.385",
    pages = "4692--4700",
    abstract = "During the fine-tuning phase of transfer learning, the pretrained vocabulary remains unchanged, while model parameters are updated. The vocabulary generated based on the pretrained data is suboptimal for downstream data when domain discrepancy exists. We propose to consider the vocabulary as an optimizable parameter, allowing us to update the vocabulary by expanding it with domain specific vocabulary based on a tokenization statistic. Furthermore, we preserve the embeddings of the added words from overfitting to downstream data by utilizing knowledge learned from a pretrained language model with a regularization term. Our method achieved consistent performance improvements on diverse domains (i.e., biomedical, computer science, news, and reviews)."
}