Canada Spends Logo

Explorez notre analyse du budget 2026

Grant: Alliance Grants

Queen's University

$25,000.00

Planning for Narration: How Symbolic Reasoning and LLMs Can Work Together to Tell a Story

EF 2024-25

Résumé
Telling a compelling and coherent story is a hallmark of the intelligence required for effective communication. A narrative can teach, persuade, and establish trust, but only if it is done effectively. This makes the task of narration planning (or automated storytelling) a compelling and challenging area of research. This proposal aims to build a collaboration that directly focuses on narration planning, building on state-of-the-art techniques in both the planning and deep learning areas of research. We will explore three key challenges in the area of automated narrative planning: (1) how to identify the mechanics of a narrative (i.e., the actions that are taking place); (2) how to represent and reason with the mechanics of different forms of narration; and (3) how to produce a compelling and coherent narrative from a set of given mechanics. The three challenges are tightly coupled, and we expect to contribute significantly to all three simultaneously. Automated planning models sequential decision-making problems and has a rich history in automated storytelling. We aim to build coherent narrative structures by combining the symbolic methods of automated planning with the generative capabilities of large language models (LLMs). While LLMs produce compelling rhetoric, they often lack long-term coherence. Our goal is to tightly integrate these approaches to achieve narrations that are both coherent and compelling. This project will establish our lab (and Canada broadly) as a global leader in the area of automated narration. The application areas for the work in interactive game design and training scenario creation will be actively pursued through existing collaborations, and this will have an immediate impact on those areas of research.
Programme
Alliance Grants
Bénéficiaire
Emplacement
Ontario, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
NSERC's Awards Data

Tel que publié

Application id
597890-2024
Name nom
Muise, Christian CJ
Department département
Computing, School of
Organization id
29
Institution établissement
Queen's University
Province en
Ontario
Province fr
Ontario
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2024
Competition year année de concours
2024
Award amount
25000
Program id
ALLRP
Program name en
Alliance Grants
Program name fr
Subventions Alliance
Group en
RESEARCH PARTNERSHIPS
Group fr
PARTENARIATS DE RECHERCHE
Committee code
96
Committee name en
RPP Internal Decision Cttee
Committee name fr
Cté de décision interne - PPR
Area of application code
1200
Area of application group en
Advancement of knowledge
Area of application group fr
Avancement des connaissances
Area of application en
Advancement of knowledge
Area of application fr
Avancement des connaissances
Research subject code
2800
Research subject group en
Artificial intelligence
Research subject group fr
Intelligence artificielle
Research subject en
Artificial intelligence (computer vision, use2603)
Research subject fr
Intelligence artificielle (vision artificielle, utiliser 2603)
Field of research list names en
Not available
Field of research list names fr
Non disponible
Socioeconomic objective list names en
Not available
Socioeconomic objective list names fr
Non disponible
Application title
Planning for Narration: How Symbolic Reasoning and LLMs Can Work Together to Tell a Story
Keywords
automated planning; automated storytelling; neuro-symbolic
Application summary
Telling a compelling and coherent story is a hallmark of the intelligence required for effective communication. A narrative can teach, persuade, and establish trust, but only if it is done effectively. This makes the task of narration planning (or automated storytelling) a compelling and challenging area of research. This proposal aims to build a collaboration that directly focuses on narration planning, building on state-of-the-art techniques in both the planning and deep learning areas of research. We will explore three key challenges in the area of automated narrative planning: (1) how to identify the mechanics of a narrative (i.e., the actions that are taking place); (2) how to represent and reason with the mechanics of different forms of narration; and (3) how to produce a compelling and coherent narrative from a set of given mechanics. The three challenges are tightly coupled, and we expect to contribute significantly to all three simultaneously. Automated planning models sequential decision-making problems and has a rich history in automated storytelling. We aim to build coherent narrative structures by combining the symbolic methods of automated planning with the generative capabilities of large language models (LLMs). While LLMs produce compelling rhetoric, they often lack long-term coherence. Our goal is to tightly integrate these approaches to achieve narrations that are both coherent and compelling. This project will establish our lab (and Canada broadly) as a global leader in the area of automated narration. The application areas for the work in interactive game design and training scenario creation will be actively pursued through existing collaborations, and this will have an immediate impact on those areas of research.