@inproceedings{keleg-etal-2022-automatically,
title = "Automatically Discarding Straplines to Improve Data Quality for Abstractive News Summarization",
author = "Keleg, Amr and
Lindemann, Matthias and
Liu, Danyang and
Long, Wanqiu and
Webber, Bonnie L.",
editor = "Shavrina, Tatiana and
Mikhailov, Vladislav and
Malykh, Valentin and
Artemova, Ekaterina and
Serikov, Oleg and
Protasov, Vitaly",
booktitle = "Proceedings of NLP Power! The First Workshop on Efficient Benchmarking in NLP",
month = may,
year = "2022",
address = "Dublin, Ireland",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2022.nlppower-1.5/",
doi = "10.18653/v1/2022.nlppower-1.5",
pages = "42--51",
abstract = "Recent improvements in automatic news summarization fundamentally rely on large corpora of news articles and their summaries. These corpora are often constructed by scraping news websites, which results in including not only summaries but also other kinds of texts. Apart from more generic noise, we identify straplines as a form of text scraped from news websites that commonly turn out not to be summaries. The presence of these non-summaries threatens the validity of scraped corpora as benchmarks for news summarization. We have annotated extracts from two news sources that form part of the Newsroom corpus (Grusky et al., 2018), labeling those which were straplines, those which were summaries, and those which were both. We present a rule-based strapline detection method that achieves good performance on a manually annotated test set. Automatic evaluation indicates that removing straplines and noise from the training data of a news summarizer results in higher quality summaries, with improvements as high as 7 points ROUGE score."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="keleg-etal-2022-automatically">
<titleInfo>
<title>Automatically Discarding Straplines to Improve Data Quality for Abstractive News Summarization</title>
</titleInfo>
<name type="personal">
<namePart type="given">Amr</namePart>
<namePart type="family">Keleg</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Matthias</namePart>
<namePart type="family">Lindemann</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Danyang</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Wanqiu</namePart>
<namePart type="family">Long</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Bonnie</namePart>
<namePart type="given">L</namePart>
<namePart type="family">Webber</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2022-05</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of NLP Power! The First Workshop on Efficient Benchmarking in NLP</title>
</titleInfo>
<name type="personal">
<namePart type="given">Tatiana</namePart>
<namePart type="family">Shavrina</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Vladislav</namePart>
<namePart type="family">Mikhailov</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Valentin</namePart>
<namePart type="family">Malykh</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ekaterina</namePart>
<namePart type="family">Artemova</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Oleg</namePart>
<namePart type="family">Serikov</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Vitaly</namePart>
<namePart type="family">Protasov</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Dublin, Ireland</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Recent improvements in automatic news summarization fundamentally rely on large corpora of news articles and their summaries. These corpora are often constructed by scraping news websites, which results in including not only summaries but also other kinds of texts. Apart from more generic noise, we identify straplines as a form of text scraped from news websites that commonly turn out not to be summaries. The presence of these non-summaries threatens the validity of scraped corpora as benchmarks for news summarization. We have annotated extracts from two news sources that form part of the Newsroom corpus (Grusky et al., 2018), labeling those which were straplines, those which were summaries, and those which were both. We present a rule-based strapline detection method that achieves good performance on a manually annotated test set. Automatic evaluation indicates that removing straplines and noise from the training data of a news summarizer results in higher quality summaries, with improvements as high as 7 points ROUGE score.</abstract>
<identifier type="citekey">keleg-etal-2022-automatically</identifier>
<identifier type="doi">10.18653/v1/2022.nlppower-1.5</identifier>
<location>
<url>https://aclanthology.org/2022.nlppower-1.5/</url>
</location>
<part>
<date>2022-05</date>
<extent unit="page">
<start>42</start>
<end>51</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Automatically Discarding Straplines to Improve Data Quality for Abstractive News Summarization
%A Keleg, Amr
%A Lindemann, Matthias
%A Liu, Danyang
%A Long, Wanqiu
%A Webber, Bonnie L.
%Y Shavrina, Tatiana
%Y Mikhailov, Vladislav
%Y Malykh, Valentin
%Y Artemova, Ekaterina
%Y Serikov, Oleg
%Y Protasov, Vitaly
%S Proceedings of NLP Power! The First Workshop on Efficient Benchmarking in NLP
%D 2022
%8 May
%I Association for Computational Linguistics
%C Dublin, Ireland
%F keleg-etal-2022-automatically
%X Recent improvements in automatic news summarization fundamentally rely on large corpora of news articles and their summaries. These corpora are often constructed by scraping news websites, which results in including not only summaries but also other kinds of texts. Apart from more generic noise, we identify straplines as a form of text scraped from news websites that commonly turn out not to be summaries. The presence of these non-summaries threatens the validity of scraped corpora as benchmarks for news summarization. We have annotated extracts from two news sources that form part of the Newsroom corpus (Grusky et al., 2018), labeling those which were straplines, those which were summaries, and those which were both. We present a rule-based strapline detection method that achieves good performance on a manually annotated test set. Automatic evaluation indicates that removing straplines and noise from the training data of a news summarizer results in higher quality summaries, with improvements as high as 7 points ROUGE score.
%R 10.18653/v1/2022.nlppower-1.5
%U https://aclanthology.org/2022.nlppower-1.5/
%U https://doi.org/10.18653/v1/2022.nlppower-1.5
%P 42-51
Markdown (Informal)
[Automatically Discarding Straplines to Improve Data Quality for Abstractive News Summarization](https://aclanthology.org/2022.nlppower-1.5/) (Keleg et al., nlppower 2022)
ACL