@inproceedings{boldsen-etal-2022-interpreting,
    title = "Interpreting Character Embeddings With Perceptual Representations: The Case of Shape, Sound, and Color",
    author = "Boldsen, Sidsel  and
      Agirrezabal, Manex  and
      Hollenstein, Nora",
    editor = "Muresan, Smaranda  and
      Nakov, Preslav  and
      Villavicencio, Aline",
    booktitle = "Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
    month = may,
    year = "2022",
    address = "Dublin, Ireland",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/jlcl-multiple-ingestion/2022.acl-long.470/",
    doi = "10.18653/v1/2022.acl-long.470",
    pages = "6819--6836",
    abstract = "Character-level information is included in many NLP models, but evaluating the information encoded in character representations is an open issue. We leverage perceptual representations in the form of shape, sound, and color embeddings and perform a representational similarity analysis to evaluate their correlation with textual representations in five languages. This cross-lingual analysis shows that textual character representations correlate strongly with sound representations for languages using an alphabetic script, while shape correlates with featural scripts. We further develop a set of probing classifiers to intrinsically evaluate what phonological information is encoded in character embeddings. Our results suggest that information on features such as voicing are embedded in both LSTM and transformer-based representations."
}