This paper presents the Candide model as a computational architecture for modelling human-like, narrative-based language understanding. The model starts from the idea that narratives emerge through the process of interpreting novel linguistic observations, such as utterances, paragraphs and texts, with respect to previously acquired knowledge and beliefs. Narratives are personal, as they are rooted in past experiences, and constitute perspectives on the world that might motivate different interpretations of the same observations. Concretely, the Candide model operationalises this idea by dynamically modelling the belief systems and background knowledge of individual agents, updating these as new linguistic observations come in, and exposing them to a logic reasoning engine that reveals the possible sources of divergent interpretations. Apart from introducing the foundational ideas, we also present a proof-of-concept implementation that demonstrates the approach through a number of illustrative examples.
Originele taal-2English
TitelProceedings of the The 5th Workshop on Narrative Understanding
RedacteurenNader Akoury, Elizabeth Clark, Mohit Iyyer, Snigdha Chaturvedi, Faeze Brahman, Khyathi Raghavi Chandu
UitgeverijAssociation for Computational Linguistics
Aantal pagina's10
ISBN van elektronische versie9781959429920
StatusPublished - jul 2023
Evenement5th Workshop on Narrative Understanding - Toronto, Canada
Duur: 14 jul 202314 jul 2023

Publicatie series

NaamProceedings of the Annual Meeting of the Association for Computational Linguistics
ISSN van geprinte versie0736-587X


Workshop5th Workshop on Narrative Understanding
Verkorte titelWNU
Internet adres

Bibliografische nota

Publisher Copyright:
© 2023 Association for Computational Linguistics.


Duik in de onderzoeksthema's van 'The Candide model: How narratives emerge where observations meet beliefs'. Samen vormen ze een unieke vingerafdruk.

Citeer dit