<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="bellune-etal-2026-enhancing">
    <titleInfo>
        <title>Enhancing Cultural Awareness for Haitian Creole Machine Translation</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Tabitha</namePart>
        <namePart type="family">Bellune</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Ngoc</namePart>
        <namePart type="given">Tan</namePart>
        <namePart type="family">Le</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Fatiha</namePart>
        <namePart type="family">Sadat</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2026-10</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the Eleventh Conference on Machine Translation</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Barry</namePart>
            <namePart type="family">Haddow</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Tom</namePart>
            <namePart type="family">Kocmi</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Philipp</namePart>
            <namePart type="family">Koehn</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Christof</namePart>
            <namePart type="family">Monz</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Budapest, Hungary</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
        <identifier type="isbn">979-8-89176-474-3</identifier>
    </relatedItem>
    <abstract>Natural Language Processing for low-resource languages faces a major challenge: the lack of structured and culturally grounded data. This article presents our contribution to the machine translation task from Haitian Creole to English/French. Our hybrid approach relies on the creation of an original parallel corpus of 19,148 sentence pairs, derived primarily from 55 hours of transcriptions from the Atlas Linguistique d’Haïti, combined with the synthetic generation of code-switching data. By comparing a Transformer model trained from scratch with the fine-tuning of the pre-trained NLLB- 200 multilingual model, we demonstrate that cultural adaptation is crucial. Our best system records a dramatic increase in its BLEU score of 53.61% on the cultural domain and achieves an average CSI-Match score of 76.99% in the direction of hat-eng, and 30.69% BLEU and and 73.56% in terms of average CSI-Match in the direction of hat-fra respectively, proving its capacity to accurately translate complex idiomatic expressions.</abstract>
    <identifier type="citekey">bellune-etal-2026-enhancing</identifier>
    <location>
        <url>https://aclanthology.org/2026.wmt-main.186/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>2414</start>
            <end>2420</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Enhancing Cultural Awareness for Haitian Creole Machine Translation
%A Bellune, Tabitha
%A Le, Ngoc Tan
%A Sadat, Fatiha
%Y Haddow, Barry
%Y Kocmi, Tom
%Y Koehn, Philipp
%Y Monz, Christof
%S Proceedings of the Eleventh Conference on Machine Translation
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-474-3
%F bellune-etal-2026-enhancing
%X Natural Language Processing for low-resource languages faces a major challenge: the lack of structured and culturally grounded data. This article presents our contribution to the machine translation task from Haitian Creole to English/French. Our hybrid approach relies on the creation of an original parallel corpus of 19,148 sentence pairs, derived primarily from 55 hours of transcriptions from the Atlas Linguistique d’Haïti, combined with the synthetic generation of code-switching data. By comparing a Transformer model trained from scratch with the fine-tuning of the pre-trained NLLB- 200 multilingual model, we demonstrate that cultural adaptation is crucial. Our best system records a dramatic increase in its BLEU score of 53.61% on the cultural domain and achieves an average CSI-Match score of 76.99% in the direction of hat-eng, and 30.69% BLEU and and 73.56% in terms of average CSI-Match in the direction of hat-fra respectively, proving its capacity to accurately translate complex idiomatic expressions.
%U https://aclanthology.org/2026.wmt-main.186/
%P 2414-2420

Markdown (Informal)

[Enhancing Cultural Awareness for Haitian Creole Machine Translation](https://aclanthology.org/2026.wmt-main.186/) (Bellune et al., WMT 2026)

• Enhancing Cultural Awareness for Haitian Creole Machine Translation (Bellune et al., WMT 2026)

ACL

• Tabitha Bellune, Ngoc Tan Le, and Fatiha Sadat. 2026. Enhancing Cultural Awareness for Haitian Creole Machine Translation. In Proceedings of the Eleventh Conference on Machine Translation, pages 2414–2420, Budapest, Hungary. Association for Computational Linguistics.