@inproceedings{yang-2022-mask,
title = "Mask and Regenerate: A Classifier-based Approach for Unpaired Sentiment Transformation of Reviews for Electronic Commerce Websites.",
author = "Yang, Shuo",
editor = "Ku, Lun-Wei and
Li, Cheng-Te and
Tsai, Yu-Che and
Wang, Wei-Yao",
booktitle = "Proceedings of the Tenth International Workshop on Natural Language Processing for Social Media",
month = jul,
year = "2022",
address = "Seattle, Washington",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2022.socialnlp-1.1",
doi = "10.18653/v1/2022.socialnlp-1.1",
pages = "1--10",
abstract = "Style transfer is the task of transferring a sentence into the target style while keeping its content. The major challenge is that parallel corpora are not available for various domains. In this paper, we propose a Mask-And-Regenerate approach (MAR). It learns from unpaired sentences by modifying the word-level style attributes. We cautiously integrate the deletion, insertion and substitution operations into our model. This enables our model to automatically apply different edit operations for different sentences. Specifically, we train a multilayer perceptron (MLP) as a style classifier to find out and mask style-characteristic words in the source inputs. Then we learn a language model on non-parallel data sets to score sentences and remove unnecessary masks. Finally, the masked source sentences are input to a Transformer to perform style transfer. The final results show that our proposed model exceeds baselines by about 2 per cent of accuracy for both sentiment and style transfer tasks with comparable or better content retention.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="yang-2022-mask">
<titleInfo>
<title>Mask and Regenerate: A Classifier-based Approach for Unpaired Sentiment Transformation of Reviews for Electronic Commerce Websites.</title>
</titleInfo>
<name type="personal">
<namePart type="given">Shuo</namePart>
<namePart type="family">Yang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2022-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the Tenth International Workshop on Natural Language Processing for Social Media</title>
</titleInfo>
<name type="personal">
<namePart type="given">Lun-Wei</namePart>
<namePart type="family">Ku</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Cheng-Te</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yu-Che</namePart>
<namePart type="family">Tsai</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Wei-Yao</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Seattle, Washington</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Style transfer is the task of transferring a sentence into the target style while keeping its content. The major challenge is that parallel corpora are not available for various domains. In this paper, we propose a Mask-And-Regenerate approach (MAR). It learns from unpaired sentences by modifying the word-level style attributes. We cautiously integrate the deletion, insertion and substitution operations into our model. This enables our model to automatically apply different edit operations for different sentences. Specifically, we train a multilayer perceptron (MLP) as a style classifier to find out and mask style-characteristic words in the source inputs. Then we learn a language model on non-parallel data sets to score sentences and remove unnecessary masks. Finally, the masked source sentences are input to a Transformer to perform style transfer. The final results show that our proposed model exceeds baselines by about 2 per cent of accuracy for both sentiment and style transfer tasks with comparable or better content retention.</abstract>
<identifier type="citekey">yang-2022-mask</identifier>
<identifier type="doi">10.18653/v1/2022.socialnlp-1.1</identifier>
<location>
<url>https://aclanthology.org/2022.socialnlp-1.1</url>
</location>
<part>
<date>2022-07</date>
<extent unit="page">
<start>1</start>
<end>10</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Mask and Regenerate: A Classifier-based Approach for Unpaired Sentiment Transformation of Reviews for Electronic Commerce Websites.
%A Yang, Shuo
%Y Ku, Lun-Wei
%Y Li, Cheng-Te
%Y Tsai, Yu-Che
%Y Wang, Wei-Yao
%S Proceedings of the Tenth International Workshop on Natural Language Processing for Social Media
%D 2022
%8 July
%I Association for Computational Linguistics
%C Seattle, Washington
%F yang-2022-mask
%X Style transfer is the task of transferring a sentence into the target style while keeping its content. The major challenge is that parallel corpora are not available for various domains. In this paper, we propose a Mask-And-Regenerate approach (MAR). It learns from unpaired sentences by modifying the word-level style attributes. We cautiously integrate the deletion, insertion and substitution operations into our model. This enables our model to automatically apply different edit operations for different sentences. Specifically, we train a multilayer perceptron (MLP) as a style classifier to find out and mask style-characteristic words in the source inputs. Then we learn a language model on non-parallel data sets to score sentences and remove unnecessary masks. Finally, the masked source sentences are input to a Transformer to perform style transfer. The final results show that our proposed model exceeds baselines by about 2 per cent of accuracy for both sentiment and style transfer tasks with comparable or better content retention.
%R 10.18653/v1/2022.socialnlp-1.1
%U https://aclanthology.org/2022.socialnlp-1.1
%U https://doi.org/10.18653/v1/2022.socialnlp-1.1
%P 1-10
Markdown (Informal)
[Mask and Regenerate: A Classifier-based Approach for Unpaired Sentiment Transformation of Reviews for Electronic Commerce Websites.](https://aclanthology.org/2022.socialnlp-1.1) (Yang, SocialNLP 2022)
ACL