<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="chernyavskiy-etal-2021-correcting-texts">
    <titleInfo>
        <title>Correcting Texts Generated by Transformers using Discourse Features and Web Mining</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Alexander</namePart>
        <namePart type="family">Chernyavskiy</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Dmitry</namePart>
        <namePart type="family">Ilvovsky</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Boris</namePart>
        <namePart type="family">Galitsky</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2021-09</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the Student Research Workshop Associated with RANLP 2021</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Souhila</namePart>
            <namePart type="family">Djabri</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Dinara</namePart>
            <namePart type="family">Gimadi</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Tsvetomila</namePart>
            <namePart type="family">Mihaylova</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Ivelina</namePart>
            <namePart type="family">Nikolova-Koleva</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>INCOMA Ltd.</publisher>
            <place>
                <placeTerm type="text">Online</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
    </relatedItem>
    <abstract>Recent transformer-based approaches to NLG like GPT-2 can generate syntactically coherent original texts. However, these generated texts have serious flaws: global discourse incoherence and meaninglessness of sentences in terms of entity values. We address both of these flaws: they are independent but can be combined to generate original texts that will be both consistent and truthful. This paper presents an approach to estimate the quality of discourse structure. Empirical results confirm that the discourse structure of currently generated texts is inaccurate. We propose the research directions to correct it using discourse features during the fine-tuning procedure. The suggested approach is universal and can be applied to different languages. Apart from that, we suggest a method to correct wrong entity values based on Web Mining and text alignment.</abstract>
    <identifier type="citekey">chernyavskiy-etal-2021-correcting-texts</identifier>
    <location>
        <url>https://aclanthology.org/2021.ranlp-srw.6</url>
    </location>
    <part>
        <date>2021-09</date>
        <extent unit="page">
            <start>36</start>
            <end>43</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Correcting Texts Generated by Transformers using Discourse Features and Web Mining
%A Chernyavskiy, Alexander
%A Ilvovsky, Dmitry
%A Galitsky, Boris
%Y Djabri, Souhila
%Y Gimadi, Dinara
%Y Mihaylova, Tsvetomila
%Y Nikolova-Koleva, Ivelina
%S Proceedings of the Student Research Workshop Associated with RANLP 2021
%D 2021
%8 September
%I INCOMA Ltd.
%C Online
%F chernyavskiy-etal-2021-correcting-texts
%X Recent transformer-based approaches to NLG like GPT-2 can generate syntactically coherent original texts. However, these generated texts have serious flaws: global discourse incoherence and meaninglessness of sentences in terms of entity values. We address both of these flaws: they are independent but can be combined to generate original texts that will be both consistent and truthful. This paper presents an approach to estimate the quality of discourse structure. Empirical results confirm that the discourse structure of currently generated texts is inaccurate. We propose the research directions to correct it using discourse features during the fine-tuning procedure. The suggested approach is universal and can be applied to different languages. Apart from that, we suggest a method to correct wrong entity values based on Web Mining and text alignment.
%U https://aclanthology.org/2021.ranlp-srw.6
%P 36-43

Markdown (Informal)

[Correcting Texts Generated by Transformers using Discourse Features and Web Mining](https://aclanthology.org/2021.ranlp-srw.6) (Chernyavskiy et al., RANLP 2021)

• Correcting Texts Generated by Transformers using Discourse Features and Web Mining (Chernyavskiy et al., RANLP 2021)

ACL

• Alexander Chernyavskiy, Dmitry Ilvovsky, and Boris Galitsky. 2021. Correcting Texts Generated by Transformers using Discourse Features and Web Mining. In Proceedings of the Student Research Workshop Associated with RANLP 2021, pages 36–43, Online. INCOMA Ltd..