When Does Syntax Mediate Neural Language Model Performance? Evidence from Dropout Probes

Mycal Tucker; Tiwalayo Eisape; Peng Qian; Roger Levy; Julie Shah

NAACL 2022long10 citations

When Does Syntax Mediate Neural Language Model Performance? Evidence from Dropout Probes

Mycal Tucker, Tiwalayo Eisape, Peng Qian, Roger Levy, Julie Shah

Abstract

Recent causal probing literature reveals when language models and syntactic probes use similar representations. Such techniques may yield “false negative” causality results: models may use representations of syntax, but probes may have learned to use redundant encodings of the same syntactic information. We demonstrate that models do encode syntactic information redundantly and introduce a new probe design that guides probes to consider all syntactic information present in embeddings. Using these probes, we find evidence for the use of syntax in models where prior methods did not, allowing us to boost model performance by injecting syntactic information into representations.

BibTeX

@inproceedings{tucker-etal-2022-syntax,
    title = "When Does Syntax Mediate Neural Language Model Performance? Evidence from Dropout Probes",
    author = "Tucker, Mycal  and
      Eisape, Tiwalayo  and
      Qian, Peng  and
      Levy, Roger  and
      Shah, Julie",
    editor = "Carpuat, Marine  and
      de Marneffe, Marie-Catherine  and
      Meza Ruiz, Ivan Vladimir",
    booktitle = "Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies",
    month = jul,
    year = "2022",
    address = "Seattle, United States",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2022.naacl-main.394/",
    doi = "10.18653/v1/2022.naacl-main.394",
    pages = "5393--5408"
}