Negative Lexical Constraints in Neural Machine Translation

Josef Jon, Dusan Varis, Michal Novák, João Paulo Aires, Ondřej Bojar

Abstract
This paper explores negative lexical constraining in English to Czech neural machine translation. Negative lexical constraining is used to prohibit certain words or expressions in the translation produced by the NMT model. We compared various methods based on modifying either the decoding process or the training data. The comparison was performed on two tasks: paraphrasing and feedback-based translation refinement. We also studied how the methods “evade” the constraints, meaning that the disallowed expression is still present in the output, but in a changed form, most interestingly the case where a different surface form (for example different inflection) is produced. We propose a way to mitigate the issue through training with stemmed negative constraints, so that the ability of the model to induce different forms of a word might be used to prohibit the usage of all possible forms of the constraint. This helps to some extent, but the problem still persists in many cases.
Anthology ID:
2023.mtsummit-research.31
Volume:
Proceedings of Machine Translation Summit XIX, Vol. 1: Research Track
Month:
September
Year:
2023
Address:
Macau SAR, China
Editors:
Masao Utiyama, Rui Wang
Venue:
MTSummit
SIG:
Publisher:
Asia-Pacific Association for Machine Translation
Note:
Pages:
372–384
Language:
URL:
https://aclanthology.org/2023.mtsummit-research.31
DOI:
Bibkey:
Cite (ACL):
Josef Jon, Dusan Varis, Michal Novák, João Paulo Aires, and Ondřej Bojar. 2023. Negative Lexical Constraints in Neural Machine Translation. In Proceedings of Machine Translation Summit XIX, Vol. 1: Research Track, pages 372–384, Macau SAR, China. Asia-Pacific Association for Machine Translation.
Cite (Informal):
Negative Lexical Constraints in Neural Machine Translation (Jon et al., MTSummit 2023)
Copy Citation:
PDF:
https://preview.aclanthology.org/improve-issue-templates/2023.mtsummit-research.31.pdf
PDF Search