@inproceedings{aoyama-schneider-2022-probe,
    title = "Probe-Less Probing of {BERT}{'}s Layer-Wise Linguistic Knowledge with Masked Word Prediction",
    author = "Aoyama, Tatsuya  and
      Schneider, Nathan",
    editor = "Ippolito, Daphne  and
      Li, Liunian Harold  and
      Pacheco, Maria Leonor  and
      Chen, Danqi  and
      Xue, Nianwen",
    booktitle = "Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies: Student Research Workshop",
    month = jul,
    year = "2022",
    address = "Hybrid: Seattle, Washington + Online",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/fix-sig-urls/2022.naacl-srw.25/",
    doi = "10.18653/v1/2022.naacl-srw.25",
    pages = "195--201",
    abstract = "The current study quantitatively (and qualitatively for an illustrative purpose) analyzes BERT{'}s layer-wise masked word prediction on an English corpus, and finds that (1) the layerwise localization of linguistic knowledge primarily shown in probing studies is replicated in a behavior-based design and (2) that syntactic and semantic information is encoded at different layers for words of different syntactic categories. Hypothesizing that the above results are correlated with the number of likely potential candidates of the masked word prediction, we also investigate how the results differ for tokens within multiword expressions."
}