A large-scale multi-length headline corpus for analyzing length-constrained headline generation model evaluation

Yuta Hitomi, Yuya Taguchi, Hideaki Tamori, Ko Kikuta, Jiro Nishitoba, Naoaki Okazaki, Kentaro Inui, Manabu Okumura

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Browsing news articles on multiple devices is now possible. The lengths of news article headlines have precise upper bounds, dictated by the size of the display of the relevant device or interface. Therefore, controlling the length of headlines is essential when applying the task of headline generation to news production. However, because there is no corpus of headlines of multiple lengths for a given article, previous research on controlling output length in headline generation has not discussed whether the system outputs could be adequately evaluated without multiple references of different lengths. In this paper, we introduce two corpora, which are Japanese News Corpus (JNC) and JApanese MUlti-Length Headline Corpus (JAMUL), to confirm the validity of previous evaluation settings. The JNC provides common supervision data for headline generation. The JAMUL is a large-scale evaluation dataset for headlines of three different lengths composed by professional editors. We report new findings on these corpora; for example, although the longest length reference summary can appropriately evaluate the existing methods controlling output length, this evaluation setting has several problems.

Original languageEnglish
Title of host publicationINLG 2019 - 12th International Conference on Natural Language Generation, Proceedings of the Conference
PublisherAssociation for Computational Linguistics (ACL)
Pages333-343
Number of pages11
ISBN (Electronic)9781950737949
Publication statusPublished - 2019
Externally publishedYes
Event12th International Conference on Natural Language Generation, INLG 2019 - Tokyo, Japan
Duration: 2019 Oct 292019 Nov 1

Publication series

NameINLG 2019 - 12th International Conference on Natural Language Generation, Proceedings of the Conference

Conference

Conference12th International Conference on Natural Language Generation, INLG 2019
CountryJapan
CityTokyo
Period19/10/2919/11/1

ASJC Scopus subject areas

  • Software

Fingerprint Dive into the research topics of 'A large-scale multi-length headline corpus for analyzing length-constrained headline generation model evaluation'. Together they form a unique fingerprint.

Cite this