The clarity-brevity trade-off in generating referring expressions

Imtiaz Hussain Khan, Graeme D Ritchie, Kees Van Deemter

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

Abstract

Existing algorithms for the Generation of Referring Expressions (GRE) aim at generating descriptions that allow a hearer to identify its intended referent uniquely; the length of the expression is also considered, usually as a secondary issue. We explore the possibility of making the trade-off between these two factors more explicit, via a general cost function which scores these two aspects separately. We sketch some more complex phenomena which might be amenable to this treatment.
Original languageEnglish
Title of host publicationProceedings of 4th International Conference on Natural Language Generation (INLG2006)
PublisherACL
Pages89-91
Number of pages3
Publication statusPublished - 2006
Event4th International Natural Language Generation Conference (INLG2006) - Sydney, Australia
Duration: 15 Jul 200616 Jul 2006

Conference

Conference4th International Natural Language Generation Conference (INLG2006)
CountryAustralia
CitySydney
Period15/07/0616/07/06

Fingerprint

Cost functions

Keywords

  • Generation of Referring Expressions

Cite this

Khan, I. H., Ritchie, G. D., & Van Deemter, K. (2006). The clarity-brevity trade-off in generating referring expressions. In Proceedings of 4th International Conference on Natural Language Generation (INLG2006) (pp. 89-91). ACL.

The clarity-brevity trade-off in generating referring expressions. / Khan, Imtiaz Hussain; Ritchie, Graeme D; Van Deemter, Kees.

Proceedings of 4th International Conference on Natural Language Generation (INLG2006) . ACL, 2006. p. 89-91.

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

Khan, IH, Ritchie, GD & Van Deemter, K 2006, The clarity-brevity trade-off in generating referring expressions. in Proceedings of 4th International Conference on Natural Language Generation (INLG2006) . ACL, pp. 89-91, 4th International Natural Language Generation Conference (INLG2006), Sydney, Australia, 15/07/06.
Khan IH, Ritchie GD, Van Deemter K. The clarity-brevity trade-off in generating referring expressions. In Proceedings of 4th International Conference on Natural Language Generation (INLG2006) . ACL. 2006. p. 89-91
Khan, Imtiaz Hussain ; Ritchie, Graeme D ; Van Deemter, Kees. / The clarity-brevity trade-off in generating referring expressions. Proceedings of 4th International Conference on Natural Language Generation (INLG2006) . ACL, 2006. pp. 89-91
@inproceedings{d38d31d7c13e4daf88bd67ebc1bbf789,
title = "The clarity-brevity trade-off in generating referring expressions",
abstract = "Existing algorithms for the Generation of Referring Expressions (GRE) aim at generating descriptions that allow a hearer to identify its intended referent uniquely; the length of the expression is also considered, usually as a secondary issue. We explore the possibility of making the trade-off between these two factors more explicit, via a general cost function which scores these two aspects separately. We sketch some more complex phenomena which might be amenable to this treatment.",
keywords = "Generation of Referring Expressions",
author = "Khan, {Imtiaz Hussain} and Ritchie, {Graeme D} and {Van Deemter}, Kees",
year = "2006",
language = "English",
pages = "89--91",
booktitle = "Proceedings of 4th International Conference on Natural Language Generation (INLG2006)",
publisher = "ACL",

}

TY - GEN

T1 - The clarity-brevity trade-off in generating referring expressions

AU - Khan, Imtiaz Hussain

AU - Ritchie, Graeme D

AU - Van Deemter, Kees

PY - 2006

Y1 - 2006

N2 - Existing algorithms for the Generation of Referring Expressions (GRE) aim at generating descriptions that allow a hearer to identify its intended referent uniquely; the length of the expression is also considered, usually as a secondary issue. We explore the possibility of making the trade-off between these two factors more explicit, via a general cost function which scores these two aspects separately. We sketch some more complex phenomena which might be amenable to this treatment.

AB - Existing algorithms for the Generation of Referring Expressions (GRE) aim at generating descriptions that allow a hearer to identify its intended referent uniquely; the length of the expression is also considered, usually as a secondary issue. We explore the possibility of making the trade-off between these two factors more explicit, via a general cost function which scores these two aspects separately. We sketch some more complex phenomena which might be amenable to this treatment.

KW - Generation of Referring Expressions

M3 - Conference contribution

SP - 89

EP - 91

BT - Proceedings of 4th International Conference on Natural Language Generation (INLG2006)

PB - ACL

ER -