Neural String Edit Distance

Jindřich Libovický, Alexander Fraser


Abstract
We propose the neural string edit distance model for string-pair matching and string transduction based on learnable string edit distance. We modify the original expectation-maximization learned edit distance algorithm into a differentiable loss function, allowing us to integrate it into a neural network providing a contextual representation of the input. We evaluate on cognate detection, transliteration, and grapheme-to-phoneme conversion, and show that we can trade off between performance and interpretability in a single framework. Using contextual representations, which are difficult to interpret, we match the performance of state-of-the-art string-pair matching models. Using static embeddings and a slightly different loss function, we force interpretability, at the expense of an accuracy drop.
Anthology ID:
2022.spnlp-1.6
Volume:
Proceedings of the Sixth Workshop on Structured Prediction for NLP
Month:
May
Year:
2022
Address:
Dublin, Ireland
Venues:
ACL | spnlp
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
52–66
Language:
URL:
https://aclanthology.org/2022.spnlp-1.6
DOI:
10.18653/v1/2022.spnlp-1.6
Bibkey:
Cite (ACL):
Jindřich Libovický and Alexander Fraser. 2022. Neural String Edit Distance. In Proceedings of the Sixth Workshop on Structured Prediction for NLP, pages 52–66, Dublin, Ireland. Association for Computational Linguistics.
Cite (Informal):
Neural String Edit Distance (Libovický & Fraser, spnlp 2022)
Copy Citation:
PDF:
https://aclanthology.org/2022.spnlp-1.6.pdf
Code
 jlibovicky/neural-string-edit-distance