<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="knowles-2021-stability">
    <titleInfo>
        <title>On the Stability of System Rankings at WMT</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Rebecca</namePart>
        <namePart type="family">Knowles</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2021-nov</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the Sixth Conference on Machine Translation</title>
        </titleInfo>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Online</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
    </relatedItem>
    <abstract>The current approach to collecting human judgments of machine translation quality for the news translation task at WMT – segment rating with document context – is the most recent in a sequence of changes to WMT human annotation protocol. As these annotation protocols have changed over time, they have drifted away from some of the initial statistical assumptions underpinning them, with consequences that call the validity of WMT news task system rankings into question. In simulations based on real data, we show that the rankings can be influenced by the presence of outliers (high- or low-quality systems), resulting in different system rankings and clusterings. We also examine questions of annotation task composition and how ease or difficulty of translating different documents may influence system rankings. We provide discussion of ways to analyze these issues when considering future changes to annotation protocols.</abstract>
    <identifier type="citekey">knowles-2021-stability</identifier>
    <location>
        <url>https://aclanthology.org/2021.wmt-1.56</url>
    </location>
    <part>
        <date>2021-nov</date>
        <extent unit="page">
            <start>464</start>
            <end>477</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T On the Stability of System Rankings at WMT
%A Knowles, Rebecca
%S Proceedings of the Sixth Conference on Machine Translation
%D 2021
%8 nov
%I Association for Computational Linguistics
%C Online
%F knowles-2021-stability
%X The current approach to collecting human judgments of machine translation quality for the news translation task at WMT – segment rating with document context – is the most recent in a sequence of changes to WMT human annotation protocol. As these annotation protocols have changed over time, they have drifted away from some of the initial statistical assumptions underpinning them, with consequences that call the validity of WMT news task system rankings into question. In simulations based on real data, we show that the rankings can be influenced by the presence of outliers (high- or low-quality systems), resulting in different system rankings and clusterings. We also examine questions of annotation task composition and how ease or difficulty of translating different documents may influence system rankings. We provide discussion of ways to analyze these issues when considering future changes to annotation protocols.
%U https://aclanthology.org/2021.wmt-1.56
%P 464-477

Markdown (Informal)

[On the Stability of System Rankings at WMT](https://aclanthology.org/2021.wmt-1.56) (Knowles, WMT 2021)

• On the Stability of System Rankings at WMT (Knowles, WMT 2021)

ACL

• Rebecca Knowles. 2021. On the Stability of System Rankings at WMT. In Proceedings of the Sixth Conference on Machine Translation, pages 464–477, Online. Association for Computational Linguistics.