<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="liu-etal-2021-pretraining">
    <titleInfo>
        <title>Pretraining the Noisy Channel Model for Task-Oriented Dialogue</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Qi</namePart>
        <namePart type="family">Liu</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Lei</namePart>
        <namePart type="family">Yu</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Laura</namePart>
        <namePart type="family">Rimell</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Phil</namePart>
        <namePart type="family">Blunsom</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2021</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>Direct decoding for task-oriented dialogue is known to suffer from the explaining-away effect, manifested in models that prefer short and generic responses. Here we argue for the use of Bayes’ theorem to factorize the dialogue task into two models, the distribution of the context given the response, and the prior for the response itself. This approach, an instantiation of the noisy channel model, both mitigates the explaining-away effect and allows the principled incorporation of large pretrained models for the response prior. We present extensive experiments showing that a noisy channel model decodes better responses compared to direct decoding and that a two-stage pretraining strategy, employing both open-domain and task-oriented dialogue data, improves over randomly initialized models.</abstract>
    <identifier type="citekey">liu-etal-2021-pretraining</identifier>
    <identifier type="doi">10.1162/tacl_a_00390</identifier>
    <location>
        <url>https://aclanthology.org/2021.tacl-1.40</url>
    </location>
    <part>
        <date>2021</date>
        <detail type="volume"><number>9</number></detail>
        <extent unit="page">
            <start>657</start>
            <end>674</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Journal Article
%T Pretraining the Noisy Channel Model for Task-Oriented Dialogue
%A Liu, Qi
%A Yu, Lei
%A Rimell, Laura
%A Blunsom, Phil
%J Transactions of the Association for Computational Linguistics
%D 2021
%V 9
%I MIT Press
%C Cambridge, MA
%F liu-etal-2021-pretraining
%X Direct decoding for task-oriented dialogue is known to suffer from the explaining-away effect, manifested in models that prefer short and generic responses. Here we argue for the use of Bayes’ theorem to factorize the dialogue task into two models, the distribution of the context given the response, and the prior for the response itself. This approach, an instantiation of the noisy channel model, both mitigates the explaining-away effect and allows the principled incorporation of large pretrained models for the response prior. We present extensive experiments showing that a noisy channel model decodes better responses compared to direct decoding and that a two-stage pretraining strategy, employing both open-domain and task-oriented dialogue data, improves over randomly initialized models.
%R 10.1162/tacl_a_00390
%U https://aclanthology.org/2021.tacl-1.40
%U https://doi.org/10.1162/tacl_a_00390
%P 657-674

Markdown (Informal)

[Pretraining the Noisy Channel Model for Task-Oriented Dialogue](https://aclanthology.org/2021.tacl-1.40) (Liu et al., TACL 2021)

• Pretraining the Noisy Channel Model for Task-Oriented Dialogue (Liu et al., TACL 2021)

ACL

• Qi Liu, Lei Yu, Laura Rimell, and Phil Blunsom. 2021. Pretraining the Noisy Channel Model for Task-Oriented Dialogue. Transactions of the Association for Computational Linguistics, 9:657–674.