<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="zufle-etal-2025-comet">
    <titleInfo>
        <title>COMET-poly: Machine Translation Metric Grounded in Other Candidates</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Maike</namePart>
        <namePart type="family">Züfle</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Vilém</namePart>
        <namePart type="family">Zouhar</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Tu</namePart>
        <namePart type="given">Anh</namePart>
        <namePart type="family">Dinh</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Felipe</namePart>
        <namePart type="family">Maia Polo</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Jan</namePart>
        <namePart type="family">Niehues</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Mrinmaya</namePart>
        <namePart type="family">Sachan</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2025-11</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the Tenth 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">Suzhou, China</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
        <identifier type="isbn">979-8-89176-341-8</identifier>
    </relatedItem>
    <abstract>Automated metrics for machine translation attempt to replicate human judgment. Unlike humans, who often assess a translation in the context of multiple alternatives, these metrics typically consider only the source sentence and a single translation. This discrepancy in the evaluation setup may negatively impact the performance of automated metrics. We propose two automated metrics that incorporate additional information beyond the single translation. COMET-polycand uses alternative translations of the same source sentence to compare and contrast with the translation at hand, thereby providing a more informed assessment of its quality. COMET-polyic, inspired by retrieval-based in-context learning, takes in translations of similar source texts along with their human-labeled quality scores to guide the evaluation. We find that including a single additional translation in COMET-polycand improves the segment-level metric performance (0.079 to 0.118 Kendall’s tau-b correlation), with further gains when more translations are added. Incorporating retrieved examples in COMET-polyic yields similar improvements (0.079 to 0.116 Kendall’s tau-b correlation). We release our models publicly.</abstract>
    <identifier type="citekey">zufle-etal-2025-comet</identifier>
    <location>
        <url>https://aclanthology.org/2025.wmt-1.63/</url>
    </location>
    <part>
        <date>2025-11</date>
        <extent unit="page">
            <start>887</start>
            <end>904</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T COMET-poly: Machine Translation Metric Grounded in Other Candidates
%A Züfle, Maike
%A Zouhar, Vilém
%A Dinh, Tu Anh
%A Maia Polo, Felipe
%A Niehues, Jan
%A Sachan, Mrinmaya
%Y Haddow, Barry
%Y Kocmi, Tom
%Y Koehn, Philipp
%Y Monz, Christof
%S Proceedings of the Tenth Conference on Machine Translation
%D 2025
%8 November
%I Association for Computational Linguistics
%C Suzhou, China
%@ 979-8-89176-341-8
%F zufle-etal-2025-comet
%X Automated metrics for machine translation attempt to replicate human judgment. Unlike humans, who often assess a translation in the context of multiple alternatives, these metrics typically consider only the source sentence and a single translation. This discrepancy in the evaluation setup may negatively impact the performance of automated metrics. We propose two automated metrics that incorporate additional information beyond the single translation. COMET-polycand uses alternative translations of the same source sentence to compare and contrast with the translation at hand, thereby providing a more informed assessment of its quality. COMET-polyic, inspired by retrieval-based in-context learning, takes in translations of similar source texts along with their human-labeled quality scores to guide the evaluation. We find that including a single additional translation in COMET-polycand improves the segment-level metric performance (0.079 to 0.118 Kendall’s tau-b correlation), with further gains when more translations are added. Incorporating retrieved examples in COMET-polyic yields similar improvements (0.079 to 0.116 Kendall’s tau-b correlation). We release our models publicly.
%U https://aclanthology.org/2025.wmt-1.63/
%P 887-904

Markdown (Informal)

[COMET-poly: Machine Translation Metric Grounded in Other Candidates](https://aclanthology.org/2025.wmt-1.63/) (Züfle et al., WMT 2025)

• COMET-poly: Machine Translation Metric Grounded in Other Candidates (Züfle et al., WMT 2025)

ACL

• Maike Züfle, Vilém Zouhar, Tu Anh Dinh, Felipe Maia Polo, Jan Niehues, and Mrinmaya Sachan. 2025. COMET-poly: Machine Translation Metric Grounded in Other Candidates. In Proceedings of the Tenth Conference on Machine Translation, pages 887–904, Suzhou, China. Association for Computational Linguistics.