@inproceedings{guo-etal-2020-cyclegt,
title = "{C}ycle{GT}: Unsupervised Graph-to-Text and Text-to-Graph Generation via Cycle Training",
author = "Guo, Qipeng and
Jin, Zhijing and
Qiu, Xipeng and
Zhang, Weinan and
Wipf, David and
Zhang, Zheng",
booktitle = "Proceedings of the 3rd International Workshop on Natural Language Generation from the Semantic Web (WebNLG+)",
month = "12",
year = "2020",
address = "Dublin, Ireland (Virtual)",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2020.webnlg-1.8",
pages = "77--88",
abstract = "Two important tasks at the intersection of knowledge graphs and natural language processing are graph-to-text (G2T) and text-tograph (T2G) conversion. Due to the difficulty and high cost of data collection, the supervised data available in the two fields are usually on the magnitude of tens of thousands, for example, 18K in the WebNLG 2017 dataset after preprocessing, which is far fewer than the millions of data for other tasks such as machine translation. Consequently, deep learning models for G2T and T2G suffer largely from scarce training data. We present CycleGT, an unsupervised training method that can bootstrap from fully non-parallel graph and text data, and iteratively back translate between the two forms. Experiments on WebNLG datasets show that our unsupervised model trained on the same number of data achieves performance on par with several fully supervised models. Further experiments on the non-parallel GenWiki dataset verify that our method performs the best among unsupervised baselines. This validates our framework as an effective approach to overcome the data scarcity problem in the fields of G2T and T2G.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="guo-etal-2020-cyclegt">
<titleInfo>
<title>CycleGT: Unsupervised Graph-to-Text and Text-to-Graph Generation via Cycle Training</title>
</titleInfo>
<name type="personal">
<namePart type="given">Qipeng</namePart>
<namePart type="family">Guo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhijing</namePart>
<namePart type="family">Jin</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xipeng</namePart>
<namePart type="family">Qiu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Weinan</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">David</namePart>
<namePart type="family">Wipf</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zheng</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2020-12</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 3rd International Workshop on Natural Language Generation from the Semantic Web (WebNLG+)</title>
</titleInfo>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Dublin, Ireland (Virtual)</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Two important tasks at the intersection of knowledge graphs and natural language processing are graph-to-text (G2T) and text-tograph (T2G) conversion. Due to the difficulty and high cost of data collection, the supervised data available in the two fields are usually on the magnitude of tens of thousands, for example, 18K in the WebNLG 2017 dataset after preprocessing, which is far fewer than the millions of data for other tasks such as machine translation. Consequently, deep learning models for G2T and T2G suffer largely from scarce training data. We present CycleGT, an unsupervised training method that can bootstrap from fully non-parallel graph and text data, and iteratively back translate between the two forms. Experiments on WebNLG datasets show that our unsupervised model trained on the same number of data achieves performance on par with several fully supervised models. Further experiments on the non-parallel GenWiki dataset verify that our method performs the best among unsupervised baselines. This validates our framework as an effective approach to overcome the data scarcity problem in the fields of G2T and T2G.</abstract>
<identifier type="citekey">guo-etal-2020-cyclegt</identifier>
<location>
<url>https://aclanthology.org/2020.webnlg-1.8</url>
</location>
<part>
<date>2020-12</date>
<extent unit="page">
<start>77</start>
<end>88</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T CycleGT: Unsupervised Graph-to-Text and Text-to-Graph Generation via Cycle Training
%A Guo, Qipeng
%A Jin, Zhijing
%A Qiu, Xipeng
%A Zhang, Weinan
%A Wipf, David
%A Zhang, Zheng
%S Proceedings of the 3rd International Workshop on Natural Language Generation from the Semantic Web (WebNLG+)
%D 2020
%8 December
%I Association for Computational Linguistics
%C Dublin, Ireland (Virtual)
%F guo-etal-2020-cyclegt
%X Two important tasks at the intersection of knowledge graphs and natural language processing are graph-to-text (G2T) and text-tograph (T2G) conversion. Due to the difficulty and high cost of data collection, the supervised data available in the two fields are usually on the magnitude of tens of thousands, for example, 18K in the WebNLG 2017 dataset after preprocessing, which is far fewer than the millions of data for other tasks such as machine translation. Consequently, deep learning models for G2T and T2G suffer largely from scarce training data. We present CycleGT, an unsupervised training method that can bootstrap from fully non-parallel graph and text data, and iteratively back translate between the two forms. Experiments on WebNLG datasets show that our unsupervised model trained on the same number of data achieves performance on par with several fully supervised models. Further experiments on the non-parallel GenWiki dataset verify that our method performs the best among unsupervised baselines. This validates our framework as an effective approach to overcome the data scarcity problem in the fields of G2T and T2G.
%U https://aclanthology.org/2020.webnlg-1.8
%P 77-88
Markdown (Informal)
[CycleGT: Unsupervised Graph-to-Text and Text-to-Graph Generation via Cycle Training](https://aclanthology.org/2020.webnlg-1.8) (Guo et al., WebNLG 2020)
ACL