@inproceedings{bafna-etal-2025-dialup,
    title = "{D}ial{U}p! Modeling the Language Continuum by Adapting Models to Dialects and Dialects to Models",
    author = "Bafna, Niyati  and
      Chang, Emily  and
      Robinson, Nathaniel Romney  and
      Mortensen, David R.  and
      Murray, Kenton  and
      Yarowsky, David  and
      Sirin, Hale",
    editor = "Che, Wanxiang  and
      Nabende, Joyce  and
      Shutova, Ekaterina  and
      Pilehvar, Mohammad Taher",
    booktitle = "Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
    month = jul,
    year = "2025",
    address = "Vienna, Austria",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/acl25-workshop-ingestion/2025.acl-long.989/",
    pages = "20188--20233",
    ISBN = "979-8-89176-251-0",
    abstract = "Most of the world{'}s languages and dialects are low-resource, and lack support in mainstream machine translation (MT) models. However, many of them have a closely-related high-resource language (HRL) neighbor, and differ in linguistically regular ways from it. This underscores the importance of model robustness to dialectal variation and cross-lingual generalization to the HRL dialect continuum. We present DialUp, consisting of a training-time technique for adapting a pretrained model to dialectal data (M{--}{\ensuremath{>}}D), and an inference-time intervention adapting dialectal data to the model expertise (D{--}{\ensuremath{>}}M). M{--}{\ensuremath{>}}D induces model robustness to potentially unseen and unknown dialects by exposure to synthetic data exemplifying linguistic mechanisms of dialectal variation, whereas D{--}{\ensuremath{>}}M treats dialectal divergence for known target dialects. These methods show considerable performance gains for several dialects from four language families, and modest gains for two other language families. We also conduct feature and error analyses, which show that language varieties with low baseline MT performance are more likely to benefit from these approaches."
}