@inproceedings{de-vries-etal-2020-whats,
    title = "What`s so special about {BERT}`s layers? A closer look at the {NLP} pipeline in monolingual and multilingual models",
    author = "de Vries, Wietse  and
      van Cranenburgh, Andreas  and
      Nissim, Malvina",
    editor = "Cohn, Trevor  and
      He, Yulan  and
      Liu, Yang",
    booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2020",
    month = nov,
    year = "2020",
    address = "Online",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/jlcl-multiple-ingestion/2020.findings-emnlp.389/",
    doi = "10.18653/v1/2020.findings-emnlp.389",
    pages = "4339--4350",
    abstract = "Peeking into the inner workings of BERT has shown that its layers resemble the classical NLP pipeline, with progressively more complex tasks being concentrated in later layers. To investigate to what extent these results also hold for a language other than English, we probe a Dutch BERT-based model and the multilingual BERT model for Dutch NLP tasks. In addition, through a deeper analysis of part-of-speech tagging, we show that also within a given task, information is spread over different parts of the network and the pipeline might not be as neat as it seems. Each layer has different specialisations, so that it may be more useful to combine information from different layers, instead of selecting a single one based on the best overall performance."
}