<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="cuong-etal-2016-adapting">
    <titleInfo>
        <title>Adapting to All Domains at Once: Rewarding Domain Invariance in SMT</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Hoang</namePart>
        <namePart type="family">Cuong</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Khalil</namePart>
        <namePart type="family">Sima’an</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Ivan</namePart>
        <namePart type="family">Titov</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2016</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <genre authority="bibutilsgt">journal article</genre>
    <relatedItem type="host">
        <titleInfo>
            <title>Transactions of the Association for Computational Linguistics</title>
        </titleInfo>
        <originInfo>
            <issuance>continuing</issuance>
            <publisher>MIT Press</publisher>
            <place>
                <placeTerm type="text">Cambridge, MA</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">periodical</genre>
        <genre authority="bibutilsgt">academic journal</genre>
    </relatedItem>
    <abstract>Existing work on domain adaptation for statistical machine translation has consistently assumed access to a small sample from the test distribution (target domain) at training time. In practice, however, the target domain may not be known at training time or it may change to match user needs. In such situations, it is natural to push the system to make safer choices, giving higher preference to domain-invariant translations, which work well across domains, over risky domain-specific alternatives. We encode this intuition by (1) inducing latent subdomains from the training data only; (2) introducing features which measure how specialized phrases are to individual induced sub-domains; (3) estimating feature weights on out-of-domain data (rather than on the target domain). We conduct experiments on three language pairs and a number of different domains. We observe consistent improvements over a baseline which does not explicitly reward domain invariance.</abstract>
    <identifier type="citekey">cuong-etal-2016-adapting</identifier>
    <identifier type="doi">10.1162/tacl_a_00086</identifier>
    <location>
        <url>https://aclanthology.org/Q16-1008</url>
    </location>
    <part>
        <date>2016</date>
        <detail type="volume"><number>4</number></detail>
        <extent unit="page">
            <start>99</start>
            <end>112</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Journal Article
%T Adapting to All Domains at Once: Rewarding Domain Invariance in SMT
%A Cuong, Hoang
%A Sima’an, Khalil
%A Titov, Ivan
%J Transactions of the Association for Computational Linguistics
%D 2016
%V 4
%I MIT Press
%C Cambridge, MA
%F cuong-etal-2016-adapting
%X Existing work on domain adaptation for statistical machine translation has consistently assumed access to a small sample from the test distribution (target domain) at training time. In practice, however, the target domain may not be known at training time or it may change to match user needs. In such situations, it is natural to push the system to make safer choices, giving higher preference to domain-invariant translations, which work well across domains, over risky domain-specific alternatives. We encode this intuition by (1) inducing latent subdomains from the training data only; (2) introducing features which measure how specialized phrases are to individual induced sub-domains; (3) estimating feature weights on out-of-domain data (rather than on the target domain). We conduct experiments on three language pairs and a number of different domains. We observe consistent improvements over a baseline which does not explicitly reward domain invariance.
%R 10.1162/tacl_a_00086
%U https://aclanthology.org/Q16-1008
%U https://doi.org/10.1162/tacl_a_00086
%P 99-112

Markdown (Informal)

[Adapting to All Domains at Once: Rewarding Domain Invariance in SMT](https://aclanthology.org/Q16-1008) (Cuong et al., TACL 2016)

• Adapting to All Domains at Once: Rewarding Domain Invariance in SMT (Cuong et al., TACL 2016)

ACL

• Hoang Cuong, Khalil Sima’an, and Ivan Titov. 2016. Adapting to All Domains at Once: Rewarding Domain Invariance in SMT. Transactions of the Association for Computational Linguistics, 4:99–112.