@inproceedings{cho-etal-2021-contrastive,
title = "Contrastive Multi-document Question Generation",
author = "Cho, Woon Sang and
Zhang, Yizhe and
Rao, Sudha and
Celikyilmaz, Asli and
Xiong, Chenyan and
Gao, Jianfeng and
Wang, Mengdi and
Dolan, Bill",
editor = "Merlo, Paola and
Tiedemann, Jorg and
Tsarfaty, Reut",
booktitle = "Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics: Main Volume",
month = apr,
year = "2021",
address = "Online",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2021.eacl-main.2",
doi = "10.18653/v1/2021.eacl-main.2",
pages = "12--30",
abstract = "Multi-document question generation focuses on generating a question that covers the common aspect of multiple documents. Such a model is useful in generating clarifying options. However, a naive model trained only using the targeted ({`}positive{'}) document set may generate too generic questions that cover a larger scope than delineated by the document set. To address this challenge, we introduce the contrastive learning strategy where given {`}positive{'} and {`}negative{'} sets of documents, we generate a question that is closely related to the {`}positive{'} set but is far away from the {`}negative{'} set. This setting allows generated questions to be more specific and related to the target document set. To generate such specific questions, we propose Multi-Source Coordinated Question Generator (MSCQG), a novel framework that includes a supervised learning (SL) stage and a reinforcement learning (RL) stage. In the SL stage, a single-document question generator is trained. In the RL stage, a coordinator model is trained to find optimal attention weights to align multiple single-document generators, by optimizing a reward designed to promote specificity of generated questions. We also develop an effective auxiliary objective, named Set-induced Contrastive Regularization (SCR) that improves the coordinator{'}s contrastive learning during the RL stage. We show that our model significantly outperforms several strong baselines, as measured by automatic metrics and human evaluation. The source repository is publicly available at {`}www.github.com/woonsangcho/contrast{\_}qgen{'}.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="cho-etal-2021-contrastive">
<titleInfo>
<title>Contrastive Multi-document Question Generation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Woon</namePart>
<namePart type="given">Sang</namePart>
<namePart type="family">Cho</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yizhe</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sudha</namePart>
<namePart type="family">Rao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Asli</namePart>
<namePart type="family">Celikyilmaz</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chenyan</namePart>
<namePart type="family">Xiong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jianfeng</namePart>
<namePart type="family">Gao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mengdi</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Bill</namePart>
<namePart type="family">Dolan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2021-04</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics: Main Volume</title>
</titleInfo>
<name type="personal">
<namePart type="given">Paola</namePart>
<namePart type="family">Merlo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jorg</namePart>
<namePart type="family">Tiedemann</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Reut</namePart>
<namePart type="family">Tsarfaty</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>Multi-document question generation focuses on generating a question that covers the common aspect of multiple documents. Such a model is useful in generating clarifying options. However, a naive model trained only using the targeted (‘positive’) document set may generate too generic questions that cover a larger scope than delineated by the document set. To address this challenge, we introduce the contrastive learning strategy where given ‘positive’ and ‘negative’ sets of documents, we generate a question that is closely related to the ‘positive’ set but is far away from the ‘negative’ set. This setting allows generated questions to be more specific and related to the target document set. To generate such specific questions, we propose Multi-Source Coordinated Question Generator (MSCQG), a novel framework that includes a supervised learning (SL) stage and a reinforcement learning (RL) stage. In the SL stage, a single-document question generator is trained. In the RL stage, a coordinator model is trained to find optimal attention weights to align multiple single-document generators, by optimizing a reward designed to promote specificity of generated questions. We also develop an effective auxiliary objective, named Set-induced Contrastive Regularization (SCR) that improves the coordinator’s contrastive learning during the RL stage. We show that our model significantly outperforms several strong baselines, as measured by automatic metrics and human evaluation. The source repository is publicly available at ‘www.github.com/woonsangcho/contrast_qgen’.</abstract>
<identifier type="citekey">cho-etal-2021-contrastive</identifier>
<identifier type="doi">10.18653/v1/2021.eacl-main.2</identifier>
<location>
<url>https://aclanthology.org/2021.eacl-main.2</url>
</location>
<part>
<date>2021-04</date>
<extent unit="page">
<start>12</start>
<end>30</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Contrastive Multi-document Question Generation
%A Cho, Woon Sang
%A Zhang, Yizhe
%A Rao, Sudha
%A Celikyilmaz, Asli
%A Xiong, Chenyan
%A Gao, Jianfeng
%A Wang, Mengdi
%A Dolan, Bill
%Y Merlo, Paola
%Y Tiedemann, Jorg
%Y Tsarfaty, Reut
%S Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics: Main Volume
%D 2021
%8 April
%I Association for Computational Linguistics
%C Online
%F cho-etal-2021-contrastive
%X Multi-document question generation focuses on generating a question that covers the common aspect of multiple documents. Such a model is useful in generating clarifying options. However, a naive model trained only using the targeted (‘positive’) document set may generate too generic questions that cover a larger scope than delineated by the document set. To address this challenge, we introduce the contrastive learning strategy where given ‘positive’ and ‘negative’ sets of documents, we generate a question that is closely related to the ‘positive’ set but is far away from the ‘negative’ set. This setting allows generated questions to be more specific and related to the target document set. To generate such specific questions, we propose Multi-Source Coordinated Question Generator (MSCQG), a novel framework that includes a supervised learning (SL) stage and a reinforcement learning (RL) stage. In the SL stage, a single-document question generator is trained. In the RL stage, a coordinator model is trained to find optimal attention weights to align multiple single-document generators, by optimizing a reward designed to promote specificity of generated questions. We also develop an effective auxiliary objective, named Set-induced Contrastive Regularization (SCR) that improves the coordinator’s contrastive learning during the RL stage. We show that our model significantly outperforms several strong baselines, as measured by automatic metrics and human evaluation. The source repository is publicly available at ‘www.github.com/woonsangcho/contrast_qgen’.
%R 10.18653/v1/2021.eacl-main.2
%U https://aclanthology.org/2021.eacl-main.2
%U https://doi.org/10.18653/v1/2021.eacl-main.2
%P 12-30
Markdown (Informal)
[Contrastive Multi-document Question Generation](https://aclanthology.org/2021.eacl-main.2) (Cho et al., EACL 2021)
ACL
- Woon Sang Cho, Yizhe Zhang, Sudha Rao, Asli Celikyilmaz, Chenyan Xiong, Jianfeng Gao, Mengdi Wang, and Bill Dolan. 2021. Contrastive Multi-document Question Generation. In Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics: Main Volume, pages 12–30, Online. Association for Computational Linguistics.