On the Branching Bias of Syntax Extracted from Pre-trained Language Models

Huayang Li, Lemao Liu, Guoping Huang, Shuming Shi


Abstract
Many efforts have been devoted to extracting constituency trees from pre-trained language models, often proceeding in two stages: feature definition and parsing. However, this kind of methods may suffer from the branching bias issue, which will inflate the performances on languages with the same branch it biases to. In this work, we propose quantitatively measuring the branching bias by comparing the performance gap on a language and its reversed language, which is agnostic to both language models and extracting methods. Furthermore, we analyze the impacts of three factors on the branching bias, namely feature definitions, parsing algorithms, and language models. Experiments show that several existing works exhibit branching biases, and some implementations of these three factors can introduce the branching bias.
Anthology ID:
2020.findings-emnlp.401
Volume:
Findings of the Association for Computational Linguistics: EMNLP 2020
Month:
November
Year:
2020
Address:
Online
Editors:
Trevor Cohn, Yulan He, Yang Liu
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
4473–4478
Language:
URL:
https://aclanthology.org/2020.findings-emnlp.401
DOI:
10.18653/v1/2020.findings-emnlp.401
Bibkey:
Cite (ACL):
Huayang Li, Lemao Liu, Guoping Huang, and Shuming Shi. 2020. On the Branching Bias of Syntax Extracted from Pre-trained Language Models. In Findings of the Association for Computational Linguistics: EMNLP 2020, pages 4473–4478, Online. Association for Computational Linguistics.
Cite (Informal):
On the Branching Bias of Syntax Extracted from Pre-trained Language Models (Li et al., Findings 2020)
Copy Citation:
PDF:
https://aclanthology.org/2020.findings-emnlp.401.pdf