Non-deterministic attribute selection in reference production

Albert Gatt, Roger van Gompel, Emiel Krahmer, Kees van Deemter

Research output: Chapter in Book/Report/Conference proceedingPublished conference contribution


In producing identifying descriptions, speakers often
overspecify and manifest preferences for certain attributes. However, current computational models which
incorporate this observation tend not to make precise
predictions about when and how much speakers do this.
The present paper proposes and evaluates two alternative models, based on the results of a new experiment. Unlike current models, the new ones are nondeterministic and seek to make precise quantitative predictions about the extent to which speakers overspecify.
Original languageEnglish
Title of host publicationProceedings of the CogSci workshop Production of Referring Expressions
Subtitle of host publicationBridging the gap between computational, empirical and theoretical approaches to reference
Place of PublicationBoston
PublisherCognitive Science Society
Number of pages7
Publication statusPublished - Jul 2011
EventPRE-CogSci 2011 - Boston, United States
Duration: 20 Jul 201120 Jul 2011


WorkshopPRE-CogSci 2011
Abbreviated titlePRE-CogSci 2011
Country/TerritoryUnited States
Internet address

Bibliographical note

This work forms part of the project Bridging the gap between psycholinguistics and computational linguistics: the case of Referring Expressions. Albert Gatt and Emiel Krahmer were supported by the Netherlands Organization for Scientific Research (NWO).


Dive into the research topics of 'Non-deterministic attribute selection in reference production'. Together they form a unique fingerprint.

Cite this