<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="tay-etal-2020-reverse">
    <titleInfo>
        <title>Reverse Engineering Configurations of Neural Text Generation Models</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Yi</namePart>
        <namePart type="family">Tay</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Dara</namePart>
        <namePart type="family">Bahri</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Che</namePart>
        <namePart type="family">Zheng</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Clifford</namePart>
        <namePart type="family">Brunk</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Donald</namePart>
        <namePart type="family">Metzler</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Andrew</namePart>
        <namePart type="family">Tomkins</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2020-07</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Dan</namePart>
            <namePart type="family">Jurafsky</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Joyce</namePart>
            <namePart type="family">Chai</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Natalie</namePart>
            <namePart type="family">Schluter</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Joel</namePart>
            <namePart type="family">Tetreault</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Online</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
    </relatedItem>
    <abstract>Recent advances in neural text generation modeling have resulted in a number of societal concerns related to how such approaches might be used in malicious ways. It is therefore desirable to develop a deeper understanding of the fundamental properties of such models. The study of artifacts that emerge in machine generated text as a result of modeling choices is a nascent research area. To this end, the extent and degree to which these artifacts surface in generated text is still unclear. In the spirit of better understanding generative text models and their artifacts, we propose the new task of distinguishing which of several variants of a given model generated some piece of text. Specifically, we conduct an extensive suite of diagnostic tests to observe whether modeling choices (e.g., sampling methods, top-k probabilities, model architectures, etc.) leave detectable artifacts in the text they generate. Our key finding, which is backed by a rigorous set of experiments, is that such artifacts are present and that different modeling choices can be inferred by looking at generated text alone. This suggests that neural text generators may actually be more sensitive to various modeling choices than previously thought.</abstract>
    <identifier type="citekey">tay-etal-2020-reverse</identifier>
    <identifier type="doi">10.18653/v1/2020.acl-main.25</identifier>
    <location>
        <url>https://aclanthology.org/2020.acl-main.25</url>
    </location>
    <part>
        <date>2020-07</date>
        <extent unit="page">
            <start>275</start>
            <end>279</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Reverse Engineering Configurations of Neural Text Generation Models
%A Tay, Yi
%A Bahri, Dara
%A Zheng, Che
%A Brunk, Clifford
%A Metzler, Donald
%A Tomkins, Andrew
%Y Jurafsky, Dan
%Y Chai, Joyce
%Y Schluter, Natalie
%Y Tetreault, Joel
%S Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics
%D 2020
%8 July
%I Association for Computational Linguistics
%C Online
%F tay-etal-2020-reverse
%X Recent advances in neural text generation modeling have resulted in a number of societal concerns related to how such approaches might be used in malicious ways. It is therefore desirable to develop a deeper understanding of the fundamental properties of such models. The study of artifacts that emerge in machine generated text as a result of modeling choices is a nascent research area. To this end, the extent and degree to which these artifacts surface in generated text is still unclear. In the spirit of better understanding generative text models and their artifacts, we propose the new task of distinguishing which of several variants of a given model generated some piece of text. Specifically, we conduct an extensive suite of diagnostic tests to observe whether modeling choices (e.g., sampling methods, top-k probabilities, model architectures, etc.) leave detectable artifacts in the text they generate. Our key finding, which is backed by a rigorous set of experiments, is that such artifacts are present and that different modeling choices can be inferred by looking at generated text alone. This suggests that neural text generators may actually be more sensitive to various modeling choices than previously thought.
%R 10.18653/v1/2020.acl-main.25
%U https://aclanthology.org/2020.acl-main.25
%U https://doi.org/10.18653/v1/2020.acl-main.25
%P 275-279

Markdown (Informal)

[Reverse Engineering Configurations of Neural Text Generation Models](https://aclanthology.org/2020.acl-main.25) (Tay et al., ACL 2020)

• Reverse Engineering Configurations of Neural Text Generation Models (Tay et al., ACL 2020)

ACL

• Yi Tay, Dara Bahri, Che Zheng, Clifford Brunk, Donald Metzler, and Andrew Tomkins. 2020. Reverse Engineering Configurations of Neural Text Generation Models. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 275–279, Online. Association for Computational Linguistics.