WSPAlign: Word Alignment Pre-training via Large-Scale Weakly Supervised Span Prediction

Qiyu Wu, Masaaki Nagata, Yoshimasa Tsuruoka


Abstract
Most existing word alignment methods rely on manual alignment datasets or parallel corpora, which limits their usefulness. Here, to mitigate the dependence on manual data, we broaden the source of supervision by relaxing the requirement for correct, fully-aligned, and parallel sentences. Specifically, we make noisy, partially aligned, and non-parallel paragraphs in this paper. We then use such a large-scale weakly-supervised dataset for word alignment pre-training via span prediction. Extensive experiments with various settings empirically demonstrate that our approach, which is named WSPAlign, is an effective and scalable way to pre-train word aligners without manual data. When fine-tuned on standard benchmarks, WSPAlign has set a new state of the art by improving upon the best supervised baseline by 3.3 6.1 points in F1 and 1.5 6.1 points in AER. Furthermore, WSPAlign also achieves competitive performance compared with the corresponding baselines in few-shot, zero-shot and cross-lingual tests, which demonstrates that WSPAlign is potentially more practical for low-resource languages than existing methods.
Anthology ID:
2023.acl-long.621
Volume:
Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
Month:
July
Year:
2023
Address:
Toronto, Canada
Editors:
Anna Rogers, Jordan Boyd-Graber, Naoaki Okazaki
Venue:
ACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
11084–11099
Language:
URL:
https://aclanthology.org/2023.acl-long.621
DOI:
10.18653/v1/2023.acl-long.621
Bibkey:
Cite (ACL):
Qiyu Wu, Masaaki Nagata, and Yoshimasa Tsuruoka. 2023. WSPAlign: Word Alignment Pre-training via Large-Scale Weakly Supervised Span Prediction. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 11084–11099, Toronto, Canada. Association for Computational Linguistics.
Cite (Informal):
WSPAlign: Word Alignment Pre-training via Large-Scale Weakly Supervised Span Prediction (Wu et al., ACL 2023)
Copy Citation:
PDF:
https://aclanthology.org/2023.acl-long.621.pdf
Video:
 https://aclanthology.org/2023.acl-long.621.mp4