Abstraction not Memory: BERT and the English Article System

Harish Tayyar Madabushi, Dagmar Divjak, Petar Milin

Research output: Working paper/PreprintPreprint

3 Downloads (Pure)

Abstract

Article prediction is a task that has long defied accurate linguistic description. As such, this task is ideally suited to evaluate models on their ability to emulate native-speaker intuition. To this end, we compare the performance of native English speakers and pre-trained models on the task of article prediction set up as a three way choice (a/an, the, zero). Our experiments with BERT show that BERT outperforms humans on this task across all articles. In particular, BERT is far superior to humans at detecting the zero article, possibly because we insert them using rules that the deep neural model can easily pick up. More interestingly, we find that BERT tends to agree more with annotators than with the corpus when inter-annotator agreement is high but switches to agreeing more with the corpus as inter-annotator agreement drops. We contend that this alignment with annotators, despite being trained on the corpus, suggests that BERT is not memorising article use, but captures a high level generalisation of article use akin to human intuition.
Original languageEnglish
Publication statusPublished - 8 Jun 2022

Bibliographical note

Accepted for publication at 2022 Annual Conference of the North American Chapter of the Association for Computational Linguistics (NAACL 2022). Data and code available at https://github.com/H-TayyarMadabushi/Abstraction-not-Memory-BERT-and-the-English-Article-System-NAACL-2022

Keywords

  • cs.CL

Fingerprint

Dive into the research topics of 'Abstraction not Memory: BERT and the English Article System'. Together they form a unique fingerprint.

Cite this