A Closer Look at Linguistic Knowledge in Masked Language Models: The Case of Relative Clauses in American English

Marius Mosbach, Stefania Degaetano-Ortlieb, Marie-Pauline Krielke, Badr M. Abdullah, Dietrich Klakow


Abstract
Transformer-based language models achieve high performance on various tasks, but we still lack understanding of the kind of linguistic knowledge they learn and rely on. We evaluate three models (BERT, RoBERTa, and ALBERT), testing their grammatical and semantic knowledge by sentence-level probing, diagnostic cases, and masked prediction tasks. We focus on relative clauses (in American English) as a complex phenomenon needing contextual information and antecedent identification to be resolved. Based on a naturalistic dataset, probing shows that all three models indeed capture linguistic knowledge about grammaticality, achieving high performance. Evaluation on diagnostic cases and masked prediction tasks considering fine-grained linguistic knowledge, however, shows pronounced model-specific weaknesses especially on semantic knowledge, strongly impacting models’ performance. Our results highlight the importance of (a)model comparison in evaluation task and (b) building up claims of model performance and the linguistic knowledge they capture beyond purely probing-based evaluations.
Anthology ID:
2020.coling-main.67
Volume:
Proceedings of the 28th International Conference on Computational Linguistics
Month:
December
Year:
2020
Address:
Barcelona, Spain (Online)
Editors:
Donia Scott, Nuria Bel, Chengqing Zong
Venue:
COLING
SIG:
Publisher:
International Committee on Computational Linguistics
Note:
Pages:
771–787
Language:
URL:
https://aclanthology.org/2020.coling-main.67
DOI:
10.18653/v1/2020.coling-main.67
Bibkey:
Cite (ACL):
Marius Mosbach, Stefania Degaetano-Ortlieb, Marie-Pauline Krielke, Badr M. Abdullah, and Dietrich Klakow. 2020. A Closer Look at Linguistic Knowledge in Masked Language Models: The Case of Relative Clauses in American English. In Proceedings of the 28th International Conference on Computational Linguistics, pages 771–787, Barcelona, Spain (Online). International Committee on Computational Linguistics.
Cite (Informal):
A Closer Look at Linguistic Knowledge in Masked Language Models: The Case of Relative Clauses in American English (Mosbach et al., COLING 2020)
Copy Citation:
PDF:
https://aclanthology.org/2020.coling-main.67.pdf
Code
 uds-lsv/rc-probing
Data
GLUE