@inproceedings{ma-etal-2022-improving,
title = "Improving Text Simplification with Factuality Error Detection",
author = "Ma, Yuan and
Seneviratne, Sandaru and
Daskalaki, Elena",
editor = "{\v{S}}tajner, Sanja and
Saggion, Horacio and
Ferr{\'e}s, Daniel and
Shardlow, Matthew and
Sheang, Kim Cheng and
North, Kai and
Zampieri, Marcos and
Xu, Wei",
booktitle = "Proceedings of the Workshop on Text Simplification, Accessibility, and Readability (TSAR-2022)",
month = dec,
year = "2022",
address = "Abu Dhabi, United Arab Emirates (Virtual)",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2022.tsar-1.16/",
doi = "10.18653/v1/2022.tsar-1.16",
pages = "173--178",
abstract = "In the past few years, the field of text simplification has been dominated by supervised learning approaches thanks to the appearance of large parallel datasets such as Wikilarge and Newsela. However, these datasets suffer from sentence pairs with factuality errors which compromise the models' performance. So, we proposed a model-independent factuality error detection mechanism, considering bad simplification and bad alignment, to refine the Wikilarge dataset through reducing the weight of these samples during training. We demonstrated that this approach improved the performance of the state-of-the-art text simplification model TST5 by an FKGL reduction of 0.33 and 0.29 on the TurkCorpus and ASSET testing datasets respectively. Our study illustrates the impact of erroneous samples in TS datasets and highlights the need for automatic methods to improve their quality."
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="ma-etal-2022-improving">
<titleInfo>
<title>Improving Text Simplification with Factuality Error Detection</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yuan</namePart>
<namePart type="family">Ma</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sandaru</namePart>
<namePart type="family">Seneviratne</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Elena</namePart>
<namePart type="family">Daskalaki</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2022-12</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the Workshop on Text Simplification, Accessibility, and Readability (TSAR-2022)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Sanja</namePart>
<namePart type="family">Štajner</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Horacio</namePart>
<namePart type="family">Saggion</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Daniel</namePart>
<namePart type="family">Ferrés</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Matthew</namePart>
<namePart type="family">Shardlow</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kim</namePart>
<namePart type="given">Cheng</namePart>
<namePart type="family">Sheang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kai</namePart>
<namePart type="family">North</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Marcos</namePart>
<namePart type="family">Zampieri</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>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Abu Dhabi, United Arab Emirates (Virtual)</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>In the past few years, the field of text simplification has been dominated by supervised learning approaches thanks to the appearance of large parallel datasets such as Wikilarge and Newsela. However, these datasets suffer from sentence pairs with factuality errors which compromise the models’ performance. So, we proposed a model-independent factuality error detection mechanism, considering bad simplification and bad alignment, to refine the Wikilarge dataset through reducing the weight of these samples during training. We demonstrated that this approach improved the performance of the state-of-the-art text simplification model TST5 by an FKGL reduction of 0.33 and 0.29 on the TurkCorpus and ASSET testing datasets respectively. Our study illustrates the impact of erroneous samples in TS datasets and highlights the need for automatic methods to improve their quality.</abstract>
<identifier type="citekey">ma-etal-2022-improving</identifier>
<identifier type="doi">10.18653/v1/2022.tsar-1.16</identifier>
<location>
<url>https://aclanthology.org/2022.tsar-1.16/</url>
</location>
<part>
<date>2022-12</date>
<extent unit="page">
<start>173</start>
<end>178</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Improving Text Simplification with Factuality Error Detection
%A Ma, Yuan
%A Seneviratne, Sandaru
%A Daskalaki, Elena
%Y Štajner, Sanja
%Y Saggion, Horacio
%Y Ferrés, Daniel
%Y Shardlow, Matthew
%Y Sheang, Kim Cheng
%Y North, Kai
%Y Zampieri, Marcos
%Y Xu, Wei
%S Proceedings of the Workshop on Text Simplification, Accessibility, and Readability (TSAR-2022)
%D 2022
%8 December
%I Association for Computational Linguistics
%C Abu Dhabi, United Arab Emirates (Virtual)
%F ma-etal-2022-improving
%X In the past few years, the field of text simplification has been dominated by supervised learning approaches thanks to the appearance of large parallel datasets such as Wikilarge and Newsela. However, these datasets suffer from sentence pairs with factuality errors which compromise the models’ performance. So, we proposed a model-independent factuality error detection mechanism, considering bad simplification and bad alignment, to refine the Wikilarge dataset through reducing the weight of these samples during training. We demonstrated that this approach improved the performance of the state-of-the-art text simplification model TST5 by an FKGL reduction of 0.33 and 0.29 on the TurkCorpus and ASSET testing datasets respectively. Our study illustrates the impact of erroneous samples in TS datasets and highlights the need for automatic methods to improve their quality.
%R 10.18653/v1/2022.tsar-1.16
%U https://aclanthology.org/2022.tsar-1.16/
%U https://doi.org/10.18653/v1/2022.tsar-1.16
%P 173-178
Markdown (Informal)
[Improving Text Simplification with Factuality Error Detection](https://aclanthology.org/2022.tsar-1.16/) (Ma et al., TSAR 2022)
ACL
- Yuan Ma, Sandaru Seneviratne, and Elena Daskalaki. 2022. Improving Text Simplification with Factuality Error Detection. In Proceedings of the Workshop on Text Simplification, Accessibility, and Readability (TSAR-2022), pages 173–178, Abu Dhabi, United Arab Emirates (Virtual). Association for Computational Linguistics.