<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="chan-etal-2021-controllable">
    <titleInfo>
        <title>Controllable Summarization with Constrained Markov Decision Process</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Hou</namePart>
        <namePart type="given">Pong</namePart>
        <namePart type="family">Chan</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Lu</namePart>
        <namePart type="family">Wang</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Irwin</namePart>
        <namePart type="family">King</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>We study controllable text summarization, which allows users to gain control on a particular attribute (e.g., length limit) of the generated summaries. In this work, we propose a novel training framework based on Constrained Markov Decision Process (CMDP), which conveniently includes a reward function along with a set of constraints, to facilitate better summarization control. The reward function encourages the generation to resemble the human-written reference, while the constraints are used to explicitly prevent the generated summaries from violating user-imposed requirements. Our framework can be applied to control important attributes of summarization, including length, covered entities, and abstractiveness, as we devise specific constraints for each of these aspects. Extensive experiments on popular benchmarks show that our CMDP framework helps generate informative summaries while complying with a given attribute’s requirement.1</abstract>
    <identifier type="citekey">chan-etal-2021-controllable</identifier>
    <identifier type="doi">10.1162/tacl_a_00423</identifier>
    <location>
        <url>https://aclanthology.org/2021.tacl-1.72</url>
    </location>
    <part>
        <date>2021</date>
        <detail type="volume"><number>9</number></detail>
        <extent unit="page">
            <start>1213</start>
            <end>1232</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Journal Article
%T Controllable Summarization with Constrained Markov Decision Process
%A Chan, Hou Pong
%A Wang, Lu
%A King, Irwin
%J Transactions of the Association for Computational Linguistics
%D 2021
%V 9
%I MIT Press
%C Cambridge, MA
%F chan-etal-2021-controllable
%X We study controllable text summarization, which allows users to gain control on a particular attribute (e.g., length limit) of the generated summaries. In this work, we propose a novel training framework based on Constrained Markov Decision Process (CMDP), which conveniently includes a reward function along with a set of constraints, to facilitate better summarization control. The reward function encourages the generation to resemble the human-written reference, while the constraints are used to explicitly prevent the generated summaries from violating user-imposed requirements. Our framework can be applied to control important attributes of summarization, including length, covered entities, and abstractiveness, as we devise specific constraints for each of these aspects. Extensive experiments on popular benchmarks show that our CMDP framework helps generate informative summaries while complying with a given attribute’s requirement.1
%R 10.1162/tacl_a_00423
%U https://aclanthology.org/2021.tacl-1.72
%U https://doi.org/10.1162/tacl_a_00423
%P 1213-1232

Markdown (Informal)

[Controllable Summarization with Constrained Markov Decision Process](https://aclanthology.org/2021.tacl-1.72) (Chan et al., TACL 2021)

• Controllable Summarization with Constrained Markov Decision Process (Chan et al., TACL 2021)

ACL

• Hou Pong Chan, Lu Wang, and Irwin King. 2021. Controllable Summarization with Constrained Markov Decision Process. Transactions of the Association for Computational Linguistics, 9:1213–1232.