@inproceedings{arnett-etal-2024-different,
    title = "Different Tokenization Schemes Lead to Comparable Performance in {S}panish Number Agreement",
    author = "Arnett, Catherine  and
      Rivi{\`e}re, Pamela D.  and
      Chang, Tyler A.  and
      Trott, Sean",
    editor = {Nicolai, Garrett  and
      Chodroff, Eleanor  and
      Mailhot, Frederic  and
      {\c{C}}{\"o}ltekin, {\c{C}}a{\u{g}}r{\i}},
    booktitle = "Proceedings of the 21st SIGMORPHON workshop on Computational Research in Phonetics, Phonology, and Morphology",
    month = jun,
    year = "2024",
    address = "Mexico City, Mexico",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/fix-sig-urls/2024.sigmorphon-1.4/",
    doi = "10.18653/v1/2024.sigmorphon-1.4",
    pages = "32--38",
    abstract = "The relationship between language model tokenization and performance is an open area of research. Here, we investigate how different tokenization schemes impact number agreement in Spanish plurals. We find that morphologically-aligned tokenization performs similarly to other tokenization schemes, even when induced artificially for words that would not be tokenized that way during training. We then present exploratory analyses demonstrating that language model embeddings for different plural tokenizations have similar distributions along the embedding space axis that maximally distinguishes singular and plural nouns. Our results suggest that morphologically-aligned tokenization is a viable tokenization approach, and existing models already generalize some morphological patterns to new items. However, our results indicate that morphological tokenization is not strictly required for performance."
}