@inproceedings{li-etal-2021-transformers,
    title = "Are {T}ransformers a Modern Version of {ELIZA}? {O}bservations on {F}rench Object Verb Agreement",
    author = "Li, Bingzhi  and
      Wisniewski, Guillaume  and
      Crabb{\'e}, Benoit",
    editor = "Moens, Marie-Francine  and
      Huang, Xuanjing  and
      Specia, Lucia  and
      Yih, Scott Wen-tau",
    booktitle = "Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing",
    month = nov,
    year = "2021",
    address = "Online and Punta Cana, Dominican Republic",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/jlcl-multiple-ingestion/2021.emnlp-main.377/",
    doi = "10.18653/v1/2021.emnlp-main.377",
    pages = "4599--4610",
    abstract = "Many recent works have demonstrated that unsupervised sentence representations of neural networks encode syntactic information by observing that neural language models are able to predict the agreement between a verb and its subject. We take a critical look at this line of research by showing that it is possible to achieve high accuracy on this agreement task with simple surface heuristics, indicating a possible flaw in our assessment of neural networks' syntactic ability. Our fine-grained analyses of results on the long-range French object-verb agreement show that contrary to LSTMs, Transformers are able to capture a non-trivial amount of grammatical structure."
}