Causal Analysis of Syntactic Agreement Mechanisms in Neural Language Models

Matthew Finlayson, Aaron Mueller, Sebastian Gehrmann, Stuart Shieber, Tal Linzen, Yonatan Belinkov


Abstract
Targeted syntactic evaluations have demonstrated the ability of language models to perform subject-verb agreement given difficult contexts. To elucidate the mechanisms by which the models accomplish this behavior, this study applies causal mediation analysis to pre-trained neural language models. We investigate the magnitude of models’ preferences for grammatical inflections, as well as whether neurons process subject-verb agreement similarly across sentences with different syntactic structures. We uncover similarities and differences across architectures and model sizes—notably, that larger models do not necessarily learn stronger preferences. We also observe two distinct mechanisms for producing subject-verb agreement depending on the syntactic structure of the input sentence. Finally, we find that language models rely on similar sets of neurons when given sentences with similar syntactic structure.
Anthology ID:
2021.acl-long.144
Volume:
Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers)
Month:
August
Year:
2021
Address:
Online
Venues:
ACL | IJCNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
1828–1843
Language:
URL:
https://aclanthology.org/2021.acl-long.144
DOI:
10.18653/v1/2021.acl-long.144
Bibkey:
Copy Citation:
PDF:
https://aclanthology.org/2021.acl-long.144.pdf