Generating Converging Narratives for Games with Large Language Models

Douglas Summers-Stay, Clare R. Voss


Abstract
We explore methods of combining the probability distributions generated by two LLM prompts in order to generate a continuation that is appropriate for both prompts at once. This is a new capability that extends the possibilities for branching and rejoining narratives in games.
Anthology ID:
2024.games-1.6
Volume:
Proceedings of the 10th Workshop on Games and Natural Language Processing @ LREC-COLING 2024
Month:
May
Year:
2024
Address:
Torino, Italia
Editors:
Chris Madge, Jon Chamberlain, Karen Fort, Udo Kruschwitz, Stephanie Lukin
Venues:
games | WS
SIG:
Publisher:
ELRA and ICCL
Note:
Pages:
43–60
Language:
URL:
https://aclanthology.org/2024.games-1.6
DOI:
Bibkey:
Cite (ACL):
Douglas Summers-Stay and Clare R. Voss. 2024. Generating Converging Narratives for Games with Large Language Models. In Proceedings of the 10th Workshop on Games and Natural Language Processing @ LREC-COLING 2024, pages 43–60, Torino, Italia. ELRA and ICCL.
Cite (Informal):
Generating Converging Narratives for Games with Large Language Models (Summers-Stay & Voss, games-WS 2024)
Copy Citation:
PDF:
https://aclanthology.org/2024.games-1.6.pdf