<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="almasi-schionning-2023-fine">
    <titleInfo>
        <title>Fine-Tuning GPT-3 for Synthetic Danish News Generation</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Mina</namePart>
        <namePart type="family">Almasi</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Anton</namePart>
        <namePart type="family">Schiønning</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2023-09</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the 16th International Natural Language Generation Conference</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">C</namePart>
            <namePart type="given">Maria</namePart>
            <namePart type="family">Keet</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Hung-Yi</namePart>
            <namePart type="family">Lee</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Sina</namePart>
            <namePart type="family">Zarrieß</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Prague, Czechia</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
    </relatedItem>
    <abstract>While GPT-3 has garnered significant attention for its capabilities in natural language generation, research on its use outside of English is still relatively limited. We focus on how GPT-3 can be fine-tuned for generating synthetic news articles in a low-resource language, namely Danish. The model’s performance is evaluated on the dimensions of human and machine detection in two separate experiments. When presented with either a real or GPT-3 generated news article, human participants achieve a 58.1% classification accuracy. Contrarily, a fine-tuned BERT classifier obtains a 92.7% accuracy on the same task. This discrepancy likely pertains to the fine-tuned GPT-3 model oversampling high-likelihood tokens in its text generation. Although this is undetectable to the human eye, it leaves a statistical discrepancy for machine classifiers to detect. We address how decisions in the experimental design favoured the machine classifiers over the human evaluators, and whether the produced synthetic articles are applicable in a real-world context.</abstract>
    <identifier type="citekey">almasi-schionning-2023-fine</identifier>
    <identifier type="doi">10.18653/v1/2023.inlg-main.4</identifier>
    <location>
        <url>https://aclanthology.org/2023.inlg-main.4</url>
    </location>
    <part>
        <date>2023-09</date>
        <extent unit="page">
            <start>54</start>
            <end>68</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Fine-Tuning GPT-3 for Synthetic Danish News Generation
%A Almasi, Mina
%A Schiønning, Anton
%Y Keet, C. Maria
%Y Lee, Hung-Yi
%Y Zarrieß, Sina
%S Proceedings of the 16th International Natural Language Generation Conference
%D 2023
%8 September
%I Association for Computational Linguistics
%C Prague, Czechia
%F almasi-schionning-2023-fine
%X While GPT-3 has garnered significant attention for its capabilities in natural language generation, research on its use outside of English is still relatively limited. We focus on how GPT-3 can be fine-tuned for generating synthetic news articles in a low-resource language, namely Danish. The model’s performance is evaluated on the dimensions of human and machine detection in two separate experiments. When presented with either a real or GPT-3 generated news article, human participants achieve a 58.1% classification accuracy. Contrarily, a fine-tuned BERT classifier obtains a 92.7% accuracy on the same task. This discrepancy likely pertains to the fine-tuned GPT-3 model oversampling high-likelihood tokens in its text generation. Although this is undetectable to the human eye, it leaves a statistical discrepancy for machine classifiers to detect. We address how decisions in the experimental design favoured the machine classifiers over the human evaluators, and whether the produced synthetic articles are applicable in a real-world context.
%R 10.18653/v1/2023.inlg-main.4
%U https://aclanthology.org/2023.inlg-main.4
%U https://doi.org/10.18653/v1/2023.inlg-main.4
%P 54-68

Markdown (Informal)

[Fine-Tuning GPT-3 for Synthetic Danish News Generation](https://aclanthology.org/2023.inlg-main.4) (Almasi & Schiønning, INLG-SIGDIAL 2023)

• Fine-Tuning GPT-3 for Synthetic Danish News Generation (Almasi & Schiønning, INLG-SIGDIAL 2023)

ACL

• Mina Almasi and Anton Schiønning. 2023. Fine-Tuning GPT-3 for Synthetic Danish News Generation. In Proceedings of the 16th International Natural Language Generation Conference, pages 54–68, Prague, Czechia. Association for Computational Linguistics.