Skip navigation

Investigating BERT’s Knowledge of Language: Five Analysis Methods with NPIs

Authors: Warstadt, Alex
Bowman, Samuel R.
et al.
Keywords: computational linguistics, natural language processing
Issue Date: Nov-2019
Publisher: Association for Computational Lingusitics
Citation: Anthology ID: D19-1286 Volume: Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP) Month: November Year: 2019 Address: Hong Kong, China Venues: EMNLP | IJCNLP SIG: SIGDAT Publisher: Association for Computational Linguistics Note: Pages: 2877–2887 URL: DOI: 10.18653/v1/D19-1286
Series/Report no.: ACL Anthology;D19-1286
Abstract: Though state-of-the-art sentence representation models can perform tasks requiring significant knowledge of grammar, it is an open question how best to evaluate their grammatical knowledge. We explore five experimental methods inspired by prior work evaluating pretrained sentence representation models. We use a single linguistic phenomenon, negative polarity item (NPI) licensing, as a case study for our experiments. NPIs like any are grammatical only if they appear in a licensing environment like negation (Sue doesn’t have any cats vs. *Sue has any cats). This phenomenon is challenging because of the variety of NPI licensing environments that exist. We introduce an artificially generated dataset that manipulates key features of NPI licensing for the experiments. We find that BERT has significant knowledge of these features, but its success varies widely across different experimental methods. We conclude that a variety of methods is necessary to reveal all relevant aspects of a model’s grammatical knowledge in a given domain.
Appears in Collections:Machine Learning for Language Lab

Files in This Item:
File Description SizeFormat 
data_generation_code.zipPython 3 code for data generation.226.92 MBUnknownView/Open
modeling_code.zipPython 3 modeling code based on the jiant package. See scripts/bert_npi for instructions.58.17 MBUnknownView/Open
npi_data_all_environments.tsvExperimental data.70.95 MBUnknownView/Open
paper.pdfResearch paper published at EMNLP 2019.1.27 MBAdobe PDFView/Open
paper_appendix.pdfAppendix/supplement to research paper published at EMNLP 2019.1.93 MBAdobe PDFView/Open

Items in FDA are protected by copyright, with all rights reserved, unless otherwise indicated.