@article{clark-etal-2022-canine,
    title = "Canine: Pre-training an Efficient Tokenization-Free Encoder for Language Representation",
    author = "Clark, Jonathan H.  and
      Garrette, Dan  and
      Turc, Iulia  and
      Wieting, John",
    editor = "Roark, Brian  and
      Nenkova, Ani",
    journal = "Transactions of the Association for Computational Linguistics",
    volume = "10",
    year = "2022",
    address = "Cambridge, MA",
    publisher = "MIT Press",
    url = "https://preview.aclanthology.org/fix-sig-urls/2022.tacl-1.5/",
    doi = "10.1162/tacl_a_00448",
    pages = "73--91",
    abstract = "Pipelined NLP systems have largely been superseded by end-to-end neural modeling, yet nearly all commonly used models still require an explicit tokenization step. While recent tokenization approaches based on data-derived subword lexicons are less brittle than manually engineered tokenizers, these techniques are not equally suited to all languages, and the use of any fixed vocabulary may limit a model{'}s ability to adapt. In this paper, we present Canine, a neural encoder that operates directly on character sequences{---}without explicit tokenization or vocabulary{---}and a pre-training strategy that operates either directly on characters or optionally uses subwords as a soft inductive bias. To use its finer-grained input effectively and efficiently, Canine combines downsampling, which reduces the input sequence length, with a deep transformer stack, which encodes context. Canine outperforms a comparable mBert model by 5.7 F1 on TyDi QA, a challenging multilingual benchmark, despite having fewer model parameters."
}