|Titre||Are Transformers a Modern Version of ELIZA? Observations on French Object Verb Agreement|
|Publication Type||Article dans des actes|
|Année de la conférence||2021|
|Authors||Li, Bingzhi, Guillaume Wisniewski, and Benoit Crabbé|
|Nom de la conférence||Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing|
|Publisher||Association for Computational Linguistics|
|Conference Location||Online and Punta Cana, Dominican Republic|
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.