@inproceedings{katinskaia-yangarber-2023-grammatical,
title = "Grammatical Error Correction for Sentence-level Assessment in Language Learning",
author = "Katinskaia, Anisia and
Yangarber, Roman",
editor = {Kochmar, Ekaterina and
Burstein, Jill and
Horbach, Andrea and
Laarmann-Quante, Ronja and
Madnani, Nitin and
Tack, Ana{\"\i}s and
Yaneva, Victoria and
Yuan, Zheng and
Zesch, Torsten},
booktitle = "Proceedings of the 18th Workshop on Innovative Use of NLP for Building Educational Applications (BEA 2023)",
month = jul,
year = "2023",
address = "Toronto, Canada",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2023.bea-1.41",
doi = "10.18653/v1/2023.bea-1.41",
pages = "488--502",
abstract = "The paper presents experiments on using a Grammatical Error Correction (GEC) model to assess the correctness of answers that language learners give to grammar exercises. We explored whether a GEC model can be applied in the language learning context for a language with complex morphology. We empirically check a hypothesis that a GEC model corrects only errors and leaves correct answers unchanged. We perform a test on assessing learner answers in a real but constrained language-learning setup: the learners answer only fill-in-the-blank and multiple-choice exercises. For this purpose, we use ReLCo, a publicly available manually annotated learner dataset in Russian (Katinskaia et al., 2022). In this experiment, we fine-tune a large-scale T5 language model for the GEC task and estimate its performance on the RULEC-GEC dataset (Rozovskaya and Roth, 2019) to compare with top-performing models. We also release an updated version of the RULEC-GEC test set, manually checked by native speakers. Our analysis shows that the GEC model performs reasonably well in detecting erroneous answers to grammar exercises and potentially can be used for best-performing error types in a real learning setup. However, it struggles to assess answers which were tagged by human annotators as alternative-correct using the aforementioned hypothesis. This is in large part due to a still low recall in correcting errors, and the fact that the GEC model may modify even correct words{---}it may generate plausible alternatives, which are hard to evaluate against the gold-standard reference.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="katinskaia-yangarber-2023-grammatical">
<titleInfo>
<title>Grammatical Error Correction for Sentence-level Assessment in Language Learning</title>
</titleInfo>
<name type="personal">
<namePart type="given">Anisia</namePart>
<namePart type="family">Katinskaia</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Roman</namePart>
<namePart type="family">Yangarber</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2023-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 18th Workshop on Innovative Use of NLP for Building Educational Applications (BEA 2023)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Ekaterina</namePart>
<namePart type="family">Kochmar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jill</namePart>
<namePart type="family">Burstein</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Andrea</namePart>
<namePart type="family">Horbach</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ronja</namePart>
<namePart type="family">Laarmann-Quante</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nitin</namePart>
<namePart type="family">Madnani</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Anaïs</namePart>
<namePart type="family">Tack</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Victoria</namePart>
<namePart type="family">Yaneva</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zheng</namePart>
<namePart type="family">Yuan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Torsten</namePart>
<namePart type="family">Zesch</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Toronto, Canada</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>The paper presents experiments on using a Grammatical Error Correction (GEC) model to assess the correctness of answers that language learners give to grammar exercises. We explored whether a GEC model can be applied in the language learning context for a language with complex morphology. We empirically check a hypothesis that a GEC model corrects only errors and leaves correct answers unchanged. We perform a test on assessing learner answers in a real but constrained language-learning setup: the learners answer only fill-in-the-blank and multiple-choice exercises. For this purpose, we use ReLCo, a publicly available manually annotated learner dataset in Russian (Katinskaia et al., 2022). In this experiment, we fine-tune a large-scale T5 language model for the GEC task and estimate its performance on the RULEC-GEC dataset (Rozovskaya and Roth, 2019) to compare with top-performing models. We also release an updated version of the RULEC-GEC test set, manually checked by native speakers. Our analysis shows that the GEC model performs reasonably well in detecting erroneous answers to grammar exercises and potentially can be used for best-performing error types in a real learning setup. However, it struggles to assess answers which were tagged by human annotators as alternative-correct using the aforementioned hypothesis. This is in large part due to a still low recall in correcting errors, and the fact that the GEC model may modify even correct words—it may generate plausible alternatives, which are hard to evaluate against the gold-standard reference.</abstract>
<identifier type="citekey">katinskaia-yangarber-2023-grammatical</identifier>
<identifier type="doi">10.18653/v1/2023.bea-1.41</identifier>
<location>
<url>https://aclanthology.org/2023.bea-1.41</url>
</location>
<part>
<date>2023-07</date>
<extent unit="page">
<start>488</start>
<end>502</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Grammatical Error Correction for Sentence-level Assessment in Language Learning
%A Katinskaia, Anisia
%A Yangarber, Roman
%Y Kochmar, Ekaterina
%Y Burstein, Jill
%Y Horbach, Andrea
%Y Laarmann-Quante, Ronja
%Y Madnani, Nitin
%Y Tack, Anaïs
%Y Yaneva, Victoria
%Y Yuan, Zheng
%Y Zesch, Torsten
%S Proceedings of the 18th Workshop on Innovative Use of NLP for Building Educational Applications (BEA 2023)
%D 2023
%8 July
%I Association for Computational Linguistics
%C Toronto, Canada
%F katinskaia-yangarber-2023-grammatical
%X The paper presents experiments on using a Grammatical Error Correction (GEC) model to assess the correctness of answers that language learners give to grammar exercises. We explored whether a GEC model can be applied in the language learning context for a language with complex morphology. We empirically check a hypothesis that a GEC model corrects only errors and leaves correct answers unchanged. We perform a test on assessing learner answers in a real but constrained language-learning setup: the learners answer only fill-in-the-blank and multiple-choice exercises. For this purpose, we use ReLCo, a publicly available manually annotated learner dataset in Russian (Katinskaia et al., 2022). In this experiment, we fine-tune a large-scale T5 language model for the GEC task and estimate its performance on the RULEC-GEC dataset (Rozovskaya and Roth, 2019) to compare with top-performing models. We also release an updated version of the RULEC-GEC test set, manually checked by native speakers. Our analysis shows that the GEC model performs reasonably well in detecting erroneous answers to grammar exercises and potentially can be used for best-performing error types in a real learning setup. However, it struggles to assess answers which were tagged by human annotators as alternative-correct using the aforementioned hypothesis. This is in large part due to a still low recall in correcting errors, and the fact that the GEC model may modify even correct words—it may generate plausible alternatives, which are hard to evaluate against the gold-standard reference.
%R 10.18653/v1/2023.bea-1.41
%U https://aclanthology.org/2023.bea-1.41
%U https://doi.org/10.18653/v1/2023.bea-1.41
%P 488-502
Markdown (Informal)
[Grammatical Error Correction for Sentence-level Assessment in Language Learning](https://aclanthology.org/2023.bea-1.41) (Katinskaia & Yangarber, BEA 2023)
ACL