@inproceedings{koyama-etal-2021-comparison,
title = "Comparison of Grammatical Error Correction Using Back-Translation Models",
author = "Koyama, Aomi and
Hotate, Kengo and
Kaneko, Masahiro and
Komachi, Mamoru",
editor = "Durmus, Esin and
Gupta, Vivek and
Liu, Nelson and
Peng, Nanyun and
Su, Yu",
booktitle = "Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Student Research Workshop",
month = jun,
year = "2021",
address = "Online",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2021.naacl-srw.16/",
doi = "10.18653/v1/2021.naacl-srw.16",
pages = "126--135",
abstract = "Grammatical error correction (GEC) suffers from a lack of sufficient parallel data. Studies on GEC have proposed several methods to generate pseudo data, which comprise pairs of grammatical and artificially produced ungrammatical sentences. Currently, a mainstream approach to generate pseudo data is back-translation (BT). Most previous studies using BT have employed the same architecture for both the GEC and BT models. However, GEC models have different correction tendencies depending on the architecture of their models. Thus, in this study, we compare the correction tendencies of GEC models trained on pseudo data generated by three BT models with different architectures, namely, Transformer, CNN, and LSTM. The results confirm that the correction tendencies for each error type are different for every BT model. In addition, we investigate the correction tendencies when using a combination of pseudo data generated by different BT models. As a result, we find that the combination of different BT models improves or interpolates the performance of each error type compared with using a single BT model with different seeds."
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="koyama-etal-2021-comparison">
<titleInfo>
<title>Comparison of Grammatical Error Correction Using Back-Translation Models</title>
</titleInfo>
<name type="personal">
<namePart type="given">Aomi</namePart>
<namePart type="family">Koyama</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kengo</namePart>
<namePart type="family">Hotate</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Masahiro</namePart>
<namePart type="family">Kaneko</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mamoru</namePart>
<namePart type="family">Komachi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2021-06</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Student Research Workshop</title>
</titleInfo>
<name type="personal">
<namePart type="given">Esin</namePart>
<namePart type="family">Durmus</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Vivek</namePart>
<namePart type="family">Gupta</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nelson</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nanyun</namePart>
<namePart type="family">Peng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yu</namePart>
<namePart type="family">Su</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>Grammatical error correction (GEC) suffers from a lack of sufficient parallel data. Studies on GEC have proposed several methods to generate pseudo data, which comprise pairs of grammatical and artificially produced ungrammatical sentences. Currently, a mainstream approach to generate pseudo data is back-translation (BT). Most previous studies using BT have employed the same architecture for both the GEC and BT models. However, GEC models have different correction tendencies depending on the architecture of their models. Thus, in this study, we compare the correction tendencies of GEC models trained on pseudo data generated by three BT models with different architectures, namely, Transformer, CNN, and LSTM. The results confirm that the correction tendencies for each error type are different for every BT model. In addition, we investigate the correction tendencies when using a combination of pseudo data generated by different BT models. As a result, we find that the combination of different BT models improves or interpolates the performance of each error type compared with using a single BT model with different seeds.</abstract>
<identifier type="citekey">koyama-etal-2021-comparison</identifier>
<identifier type="doi">10.18653/v1/2021.naacl-srw.16</identifier>
<location>
<url>https://aclanthology.org/2021.naacl-srw.16/</url>
</location>
<part>
<date>2021-06</date>
<extent unit="page">
<start>126</start>
<end>135</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Comparison of Grammatical Error Correction Using Back-Translation Models
%A Koyama, Aomi
%A Hotate, Kengo
%A Kaneko, Masahiro
%A Komachi, Mamoru
%Y Durmus, Esin
%Y Gupta, Vivek
%Y Liu, Nelson
%Y Peng, Nanyun
%Y Su, Yu
%S Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Student Research Workshop
%D 2021
%8 June
%I Association for Computational Linguistics
%C Online
%F koyama-etal-2021-comparison
%X Grammatical error correction (GEC) suffers from a lack of sufficient parallel data. Studies on GEC have proposed several methods to generate pseudo data, which comprise pairs of grammatical and artificially produced ungrammatical sentences. Currently, a mainstream approach to generate pseudo data is back-translation (BT). Most previous studies using BT have employed the same architecture for both the GEC and BT models. However, GEC models have different correction tendencies depending on the architecture of their models. Thus, in this study, we compare the correction tendencies of GEC models trained on pseudo data generated by three BT models with different architectures, namely, Transformer, CNN, and LSTM. The results confirm that the correction tendencies for each error type are different for every BT model. In addition, we investigate the correction tendencies when using a combination of pseudo data generated by different BT models. As a result, we find that the combination of different BT models improves or interpolates the performance of each error type compared with using a single BT model with different seeds.
%R 10.18653/v1/2021.naacl-srw.16
%U https://aclanthology.org/2021.naacl-srw.16/
%U https://doi.org/10.18653/v1/2021.naacl-srw.16
%P 126-135
Markdown (Informal)
[Comparison of Grammatical Error Correction Using Back-Translation Models](https://aclanthology.org/2021.naacl-srw.16/) (Koyama et al., NAACL 2021)
ACL