@inproceedings{hessel-schofield-2021-effective,
    title = "How effective is {BERT} without word ordering? Implications for language understanding and data privacy",
    author = "Hessel, Jack  and
      Schofield, Alexandra",
    editor = "Zong, Chengqing  and
      Xia, Fei  and
      Li, Wenjie  and
      Navigli, Roberto",
    booktitle = "Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 2: Short Papers)",
    month = aug,
    year = "2021",
    address = "Online",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/add-emnlp-2024-awards/2021.acl-short.27/",
    doi = "10.18653/v1/2021.acl-short.27",
    pages = "204--211",
    abstract = "Ordered word sequences contain the rich structures that define language. However, it`s often not clear if or how modern pretrained language models utilize these structures. We show that the token representations and self-attention activations within BERT are surprisingly resilient to shuffling the order of input tokens, and that for several GLUE language understanding tasks, shuffling only minimally degrades performance, e.g., by 4{\%} for QNLI. While bleak from the perspective of language understanding, our results have positive implications for cases where copyright or ethics necessitates the consideration of bag-of-words data (vs. full documents). We simulate such a scenario for three sensitive classification tasks, demonstrating minimal performance degradation vs. releasing full language sequences."
}