Creative storytelling with language models and knowledge graphs

Xinran Yang, Ilaria Tiddi

Research output: Chapter in Book / Report / Conference proceedingConference contributionAcademicpeer-review

Abstract

Automated story generation is a popular and well-recognized task in the field of natural language processing. The emergence of pre-trained language models based on large Transformer architectures shows the great capability of text generation. However, language models are limited when the generation requires explicit clues within the context. In this research, we study how to combine knowledge graphs with language models, and build a creative story generation system named DICE. DICE uses external knowledge graphs to provide context clues and implicit knowledge to generate coherent and creative stories. The evaluation shows that our approach can effectively inject the knowledge from knowledge graphs into the stories automatically generated by the language model.

Original languageEnglish
Title of host publicationCIKMW2020 Proceeding of the CIKM 2020 Workshops
Subtitle of host publicationProceedings of the CIKM 2020 Workshops co-located with 29th ACM International Conference on Information and Knowledge Management (CIKM 2020) Galway, Ireland, October 19-23, 2020
EditorsStefan Conrad, Ilaria Tiddi
PublisherCEUR-WS
Pages1-9
Number of pages9
Publication statusPublished - 15 Oct 2020
Event2020 International Conference on Information and Knowledge Management Workshops, CIKMW 2020 - Galway, Ireland
Duration: 19 Oct 202023 Oct 2020

Publication series

NameCEUR Workshop Proceedings
PublisherCEUR Workshop Proceedings
Volume2699
ISSN (Print)1613-0073

Conference

Conference2020 International Conference on Information and Knowledge Management Workshops, CIKMW 2020
Country/TerritoryIreland
CityGalway
Period19/10/2023/10/20

Keywords

  • Knowledge graph
  • Language model
  • Natural language generation
  • Story generation

Fingerprint

Dive into the research topics of 'Creative storytelling with language models and knowledge graphs'. Together they form a unique fingerprint.

Cite this