Zero-Shot Dependency Parsing with Worst-Case Aware Automated Curriculum Learning

Miryam de Lhoneux, Sheng Zhang, Anders Søgaard


Abstract
Large multilingual pretrained language models such as mBERT and XLM-RoBERTa have been found to be surprisingly effective for cross-lingual transfer of syntactic parsing models Wu and Dredze (2019), but only between related languages. However, source and training languages are rarely related, when parsing truly low-resource languages. To close this gap, we adopt a method from multi-task learning, which relies on automated curriculum learning, to dynamically optimize for parsing performance on outlier languages. We show that this approach is significantly better than uniform and size-proportional sampling in the zero-shot setting.
Anthology ID:
2022.acl-short.64
Volume:
Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers)
Month:
May
Year:
2022
Address:
Dublin, Ireland
Editors:
Smaranda Muresan, Preslav Nakov, Aline Villavicencio
Venue:
ACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
578–587
Language:
URL:
https://aclanthology.org/2022.acl-short.64
DOI:
10.18653/v1/2022.acl-short.64
Bibkey:
Cite (ACL):
Miryam de Lhoneux, Sheng Zhang, and Anders Søgaard. 2022. Zero-Shot Dependency Parsing with Worst-Case Aware Automated Curriculum Learning. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers), pages 578–587, Dublin, Ireland. Association for Computational Linguistics.
Cite (Informal):
Zero-Shot Dependency Parsing with Worst-Case Aware Automated Curriculum Learning (de Lhoneux et al., ACL 2022)
Copy Citation:
PDF:
https://aclanthology.org/2022.acl-short.64.pdf
Video:
 https://aclanthology.org/2022.acl-short.64.mp4
Code
 mdelhoneux/machamp-worst_case_acl