@inproceedings{jia-etal-2025-semantic,
title = "A Semantic Uncertainty Sampling Strategy for Back-Translation in Low-Resources Neural Machine Translation",
author = "Jia, Yepai and
Ji, Yatu and
Xue, Xiang and
Shi, Lei and
Ren, Qing-Dao-Er-Ji and
Wu, Nier and
Liu, Na and
Zhao, Chen and
Liu, Fu",
editor = "Zhao, Jin and
Wang, Mingyang and
Liu, Zhu",
booktitle = "Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 4: Student Research Workshop)",
month = jul,
year = "2025",
address = "Vienna, Austria",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.acl-srw.35/",
doi = "10.18653/v1/2025.acl-srw.35",
pages = "528--538",
ISBN = "979-8-89176-254-1",
abstract = "Back-translation has been proven effective in enhancing the performance of Neural Machine Translation (NMT), with its core mechanism relying on synthesizing parallel corpora to strengthen model training. However, while traditional back-translation methods alleviate the data scarcity in low-resource machine translation, their dependence on random sampling strategies ignores the semantic quality of monolingual data. This results in the contamination of model training through the inclusion of substantial low-quality samples in the generated corpora. To mitigate noise interference, additional training iterations or model scaling are required, significantly increasing computational costs. To address this challenge, this study proposes a Semantic Uncertainty Sampling strategy, which prioritizes sentences with higher semantic uncertainty as training samples by computationally evaluating the complexity of unannotated monolingual data. Experiments were conducted on three typical low-resource agglutinative language pairs: Mongolian-Chinese, Uyghur-Chinese, and Korean-Chinese. Results demonstrate an average BLEU score improvement of +1.7 on test sets across all three translation tasks, confirming the method{'}s effectiveness in enhancing translation accuracy and fluency. This approach provides a novel pathway for the efficient utilization of unannotated data in low-resource language scenarios."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="jia-etal-2025-semantic">
<titleInfo>
<title>A Semantic Uncertainty Sampling Strategy for Back-Translation in Low-Resources Neural Machine Translation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yepai</namePart>
<namePart type="family">Jia</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yatu</namePart>
<namePart type="family">Ji</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xiang</namePart>
<namePart type="family">Xue</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lei</namePart>
<namePart type="family">Shi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Qing-Dao-Er-Ji</namePart>
<namePart type="family">Ren</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nier</namePart>
<namePart type="family">Wu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Na</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chen</namePart>
<namePart type="family">Zhao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Fu</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 4: Student Research Workshop)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Jin</namePart>
<namePart type="family">Zhao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mingyang</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhu</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Vienna, Austria</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-254-1</identifier>
</relatedItem>
<abstract>Back-translation has been proven effective in enhancing the performance of Neural Machine Translation (NMT), with its core mechanism relying on synthesizing parallel corpora to strengthen model training. However, while traditional back-translation methods alleviate the data scarcity in low-resource machine translation, their dependence on random sampling strategies ignores the semantic quality of monolingual data. This results in the contamination of model training through the inclusion of substantial low-quality samples in the generated corpora. To mitigate noise interference, additional training iterations or model scaling are required, significantly increasing computational costs. To address this challenge, this study proposes a Semantic Uncertainty Sampling strategy, which prioritizes sentences with higher semantic uncertainty as training samples by computationally evaluating the complexity of unannotated monolingual data. Experiments were conducted on three typical low-resource agglutinative language pairs: Mongolian-Chinese, Uyghur-Chinese, and Korean-Chinese. Results demonstrate an average BLEU score improvement of +1.7 on test sets across all three translation tasks, confirming the method’s effectiveness in enhancing translation accuracy and fluency. This approach provides a novel pathway for the efficient utilization of unannotated data in low-resource language scenarios.</abstract>
<identifier type="citekey">jia-etal-2025-semantic</identifier>
<identifier type="doi">10.18653/v1/2025.acl-srw.35</identifier>
<location>
<url>https://aclanthology.org/2025.acl-srw.35/</url>
</location>
<part>
<date>2025-07</date>
<extent unit="page">
<start>528</start>
<end>538</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T A Semantic Uncertainty Sampling Strategy for Back-Translation in Low-Resources Neural Machine Translation
%A Jia, Yepai
%A Ji, Yatu
%A Xue, Xiang
%A Shi, Lei
%A Ren, Qing-Dao-Er-Ji
%A Wu, Nier
%A Liu, Na
%A Zhao, Chen
%A Liu, Fu
%Y Zhao, Jin
%Y Wang, Mingyang
%Y Liu, Zhu
%S Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 4: Student Research Workshop)
%D 2025
%8 July
%I Association for Computational Linguistics
%C Vienna, Austria
%@ 979-8-89176-254-1
%F jia-etal-2025-semantic
%X Back-translation has been proven effective in enhancing the performance of Neural Machine Translation (NMT), with its core mechanism relying on synthesizing parallel corpora to strengthen model training. However, while traditional back-translation methods alleviate the data scarcity in low-resource machine translation, their dependence on random sampling strategies ignores the semantic quality of monolingual data. This results in the contamination of model training through the inclusion of substantial low-quality samples in the generated corpora. To mitigate noise interference, additional training iterations or model scaling are required, significantly increasing computational costs. To address this challenge, this study proposes a Semantic Uncertainty Sampling strategy, which prioritizes sentences with higher semantic uncertainty as training samples by computationally evaluating the complexity of unannotated monolingual data. Experiments were conducted on three typical low-resource agglutinative language pairs: Mongolian-Chinese, Uyghur-Chinese, and Korean-Chinese. Results demonstrate an average BLEU score improvement of +1.7 on test sets across all three translation tasks, confirming the method’s effectiveness in enhancing translation accuracy and fluency. This approach provides a novel pathway for the efficient utilization of unannotated data in low-resource language scenarios.
%R 10.18653/v1/2025.acl-srw.35
%U https://aclanthology.org/2025.acl-srw.35/
%U https://doi.org/10.18653/v1/2025.acl-srw.35
%P 528-538
Markdown (Informal)
[A Semantic Uncertainty Sampling Strategy for Back-Translation in Low-Resources Neural Machine Translation](https://aclanthology.org/2025.acl-srw.35/) (Jia et al., ACL 2025)
ACL
- Yepai Jia, Yatu Ji, Xiang Xue, Lei Shi, Qing-Dao-Er-Ji Ren, Nier Wu, Na Liu, Chen Zhao, and Fu Liu. 2025. A Semantic Uncertainty Sampling Strategy for Back-Translation in Low-Resources Neural Machine Translation. In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 4: Student Research Workshop), pages 528–538, Vienna, Austria. Association for Computational Linguistics.