@inproceedings{creutz-2024-correcting,
title = "Correcting Challenging {F}innish Learner Texts With Claude, {GPT}-3.5 and {GPT}-4 Large Language Models",
author = "Creutz, Mathias",
editor = {van der Goot, Rob and
Bak, JinYeong and
M{\"u}ller-Eberstein, Max and
Xu, Wei and
Ritter, Alan and
Baldwin, Tim},
booktitle = "Proceedings of the Ninth Workshop on Noisy and User-generated Text (W-NUT 2024)",
month = mar,
year = "2024",
address = "San {\.G}iljan, Malta",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.wnut-1.1",
pages = "1--10",
abstract = "This paper studies the correction of challenging authentic Finnish learner texts at beginner level (CEFR A1). Three state-of-the-art large language models are compared, and it is shown that GPT-4 outperforms GPT-3.5, which in turn outperforms Claude v1 on this task. Additionally, ensemble models based on classifiers combining outputs of multiple single models are evaluated. The highest accuracy for an ensemble model is 84.3{\%}, whereas the best single model, which is a GPT-4 model, produces sentences that are fully correct 83.3{\%} of the time. In general, the different models perform on a continuum, where grammatical correctness, fluency and coherence go hand in hand.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="creutz-2024-correcting">
<titleInfo>
<title>Correcting Challenging Finnish Learner Texts With Claude, GPT-3.5 and GPT-4 Large Language Models</title>
</titleInfo>
<name type="personal">
<namePart type="given">Mathias</namePart>
<namePart type="family">Creutz</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2024-03</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the Ninth Workshop on Noisy and User-generated Text (W-NUT 2024)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Rob</namePart>
<namePart type="family">van der Goot</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">JinYeong</namePart>
<namePart type="family">Bak</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Max</namePart>
<namePart type="family">Müller-Eberstein</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Wei</namePart>
<namePart type="family">Xu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alan</namePart>
<namePart type="family">Ritter</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tim</namePart>
<namePart type="family">Baldwin</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">San Ġiljan, Malta</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>This paper studies the correction of challenging authentic Finnish learner texts at beginner level (CEFR A1). Three state-of-the-art large language models are compared, and it is shown that GPT-4 outperforms GPT-3.5, which in turn outperforms Claude v1 on this task. Additionally, ensemble models based on classifiers combining outputs of multiple single models are evaluated. The highest accuracy for an ensemble model is 84.3%, whereas the best single model, which is a GPT-4 model, produces sentences that are fully correct 83.3% of the time. In general, the different models perform on a continuum, where grammatical correctness, fluency and coherence go hand in hand.</abstract>
<identifier type="citekey">creutz-2024-correcting</identifier>
<location>
<url>https://aclanthology.org/2024.wnut-1.1</url>
</location>
<part>
<date>2024-03</date>
<extent unit="page">
<start>1</start>
<end>10</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Correcting Challenging Finnish Learner Texts With Claude, GPT-3.5 and GPT-4 Large Language Models
%A Creutz, Mathias
%Y van der Goot, Rob
%Y Bak, JinYeong
%Y Müller-Eberstein, Max
%Y Xu, Wei
%Y Ritter, Alan
%Y Baldwin, Tim
%S Proceedings of the Ninth Workshop on Noisy and User-generated Text (W-NUT 2024)
%D 2024
%8 March
%I Association for Computational Linguistics
%C San Ġiljan, Malta
%F creutz-2024-correcting
%X This paper studies the correction of challenging authentic Finnish learner texts at beginner level (CEFR A1). Three state-of-the-art large language models are compared, and it is shown that GPT-4 outperforms GPT-3.5, which in turn outperforms Claude v1 on this task. Additionally, ensemble models based on classifiers combining outputs of multiple single models are evaluated. The highest accuracy for an ensemble model is 84.3%, whereas the best single model, which is a GPT-4 model, produces sentences that are fully correct 83.3% of the time. In general, the different models perform on a continuum, where grammatical correctness, fluency and coherence go hand in hand.
%U https://aclanthology.org/2024.wnut-1.1
%P 1-10
Markdown (Informal)
[Correcting Challenging Finnish Learner Texts With Claude, GPT-3.5 and GPT-4 Large Language Models](https://aclanthology.org/2024.wnut-1.1) (Creutz, WNUT-WS 2024)
ACL