Skip to main navigation Skip to search Skip to main content

A large-scale Chinese long-text extractive summarization corpus

  • Harbin Institute of Technology Shenzhen

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Recently, large-scale datasets have vastly facilitated the development in nearly domains of Natural Language Processing. However, lacking large scale Chinese corpus is still a critical bottleneck for further research on deep text summarization methods. In this paper, we publish a large-scale Chinese Long-text Extractive Summarization corpus named CLES. The CLES contains about 104K <summary, article> pairs, which is originally collected from Sina Weibo1. To verify the quality of the corpus, we also manually tagged the relevance score of 5,000 <summary, article> pairs. Our benchmark models on the proposed corpus include conventional deep learning based extractive models and several pre-trained Bert-based algorithms. Their performances are reported and briefly analyzed to facilitate further research on the corpus.

Original languageEnglish
Title of host publication2021 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2021 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages7828-7832
Number of pages5
ISBN (Electronic)9781728176055
DOIs
StatePublished - 2021
Externally publishedYes
Event2021 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2021 - Virtual, Toronto, Canada
Duration: 6 Jun 202111 Jun 2021

Publication series

NameICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Volume2021-June
ISSN (Print)1520-6149

Conference

Conference2021 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2021
Country/TerritoryCanada
CityVirtual, Toronto
Period6/06/2111/06/21

Keywords

  • Large Scale
  • Long-Text
  • Pre-trained algorithm
  • Text Summarization

Fingerprint

Dive into the research topics of 'A large-scale Chinese long-text extractive summarization corpus'. Together they form a unique fingerprint.

Cite this