@inproceedings{kim-etal-2026-harnessing,
    title = "Harnessing Linguistic Dissimilarity for Language Generalization on Unseen Low-Resource Varieties",
    author = "Kim, Jinju  and
      Jung, Haeji  and
      Roh, Youjeong  and
      Ko, Jong Hwan  and
      Mortensen, David R.",
    editor = "Bonial, Claire  and
      Berzak, Yevgeni",
    booktitle = "Proceedings of the 30th Conference on Computational Natural Language Learning",
    month = jul,
    year = "2026",
    address = "San Diego, California, USA",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-acl-workshops/2026.conll-main.17/",
    pages = "284--300",
    ISBN = "979-8-89176-410-1",
    abstract = "Low-resource language varieties used by specific groups remain neglected in the development of Multilingual Language Models. A great deal of cross-lingual research focuses on inter-lingual language transfer which strives to align allied varieties and minimize differences between them. However, for low-resource varieties, linguistic dissimilarity is also an important cue allowing generalization to unseen varieties. Unlike prior approaches, we propose a two-stage Language Generalization framework that focuses on capturing variety-specific cues while also exploiting rich overlap offered by high-resource source variety. First, we propose TOPPing, a source-selection method specifically designed for low-resource varieties. Second, we suggest a lightweight VA{\c{C}}A{\'I}-Bowl architecture that learns variety-specific attributes with one branch while a parallel branch captures variety-invariant attributes using adversarial training. We evaluate our framework on structural prediction tasks, which are among the few tasks available, as proxy for performance on other downstream tasks. Using VA{\c{C}}A{\'I}-Bowl with TOPPing yields an average 54.62{\%} improvement in the dependency parsing task, which serves as a proxy for performance on other downstream tasks across 10 low-resource varieties."
}