<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="nicolai-etal-2022-penalizing">
    <titleInfo>
        <title>Penalizing Divergence: Multi-Parallel Translation for Low-Resource Languages of North America</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Garrett</namePart>
        <namePart type="family">Nicolai</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Changbing</namePart>
        <namePart type="family">Yang</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Miikka</namePart>
        <namePart type="family">Silfverberg</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2022-10</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the 29th International Conference on Computational Linguistics</title>
        </titleInfo>
        <originInfo>
            <publisher>International Committee on Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Gyeongju, Republic of Korea</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
    </relatedItem>
    <abstract>This paper explores a special case in multilingual machine translation: so called multi-parallel translation, where the target data for all language pairs are identical. While multi-parallelism offers benefits which are not available in a standard translation setting, translation models can easily overfit when training data are limited. We introduce a regularizer, the divergence penalty, which penalizes the translation model when it represents source sentences with identical target translations in divergent ways. Experiments on very low-resourced Indigenous North American languages show that an initially deficient multilingual translator can improve by 4.9 BLEU through mBART pre-training, and 5.5 BLEU points with the strategic addition of monolingual data, and that a divergence penalty leads to further increases of 0.4 BLEU. Further experiments on Germanic languages demonstrate a improvement of 0.5 BLEU when applying the divergence penalty. An investigation of the neural encoder representations learned by our translation models shows that the divergence penalty encourages models to learn a unified neural interlingua.</abstract>
    <identifier type="citekey">nicolai-etal-2022-penalizing</identifier>
    <location>
        <url>https://aclanthology.org/2022.coling-1.378</url>
    </location>
    <part>
        <date>2022-10</date>
        <extent unit="page">
            <start>4292</start>
            <end>4298</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Penalizing Divergence: Multi-Parallel Translation for Low-Resource Languages of North America
%A Nicolai, Garrett
%A Yang, Changbing
%A Silfverberg, Miikka
%S Proceedings of the 29th International Conference on Computational Linguistics
%D 2022
%8 October
%I International Committee on Computational Linguistics
%C Gyeongju, Republic of Korea
%F nicolai-etal-2022-penalizing
%X This paper explores a special case in multilingual machine translation: so called multi-parallel translation, where the target data for all language pairs are identical. While multi-parallelism offers benefits which are not available in a standard translation setting, translation models can easily overfit when training data are limited. We introduce a regularizer, the divergence penalty, which penalizes the translation model when it represents source sentences with identical target translations in divergent ways. Experiments on very low-resourced Indigenous North American languages show that an initially deficient multilingual translator can improve by 4.9 BLEU through mBART pre-training, and 5.5 BLEU points with the strategic addition of monolingual data, and that a divergence penalty leads to further increases of 0.4 BLEU. Further experiments on Germanic languages demonstrate a improvement of 0.5 BLEU when applying the divergence penalty. An investigation of the neural encoder representations learned by our translation models shows that the divergence penalty encourages models to learn a unified neural interlingua.
%U https://aclanthology.org/2022.coling-1.378
%P 4292-4298

Markdown (Informal)

[Penalizing Divergence: Multi-Parallel Translation for Low-Resource Languages of North America](https://aclanthology.org/2022.coling-1.378) (Nicolai et al., COLING 2022)

• Penalizing Divergence: Multi-Parallel Translation for Low-Resource Languages of North America (Nicolai et al., COLING 2022)

ACL

• Garrett Nicolai, Changbing Yang, and Miikka Silfverberg. 2022. Penalizing Divergence: Multi-Parallel Translation for Low-Resource Languages of North America. In Proceedings of the 29th International Conference on Computational Linguistics, pages 4292–4298, Gyeongju, Republic of Korea. International Committee on Computational Linguistics.