CoDi: A Director-Actor Framework for Goal-Driven Interactive Story Generation with LLMs
Proceedings of the AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment · 07.11.2025 · 21 (1) · 70-80
DOI: 10.1609/aiide.v21i1.36811 ↗Kurz gefasst
Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment, online erschienen am 07.11.2025. Verfasst von Honggu Kim und 2 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref).
Abstract (Originalsprache)
Interactive storytelling offers personalized and engaging narrative experiences but poses significant authoring challenges. Our proposed framework, CoDi, extends the existing director-actor paradigm by enhancing the director agent's control capabilities. Specifically, CoDi uses high-level narrative goals to facilitate adaptive storytelling, with the director agent able to introduce new events, select relevant non-player characters (NPCs), and explicitly describe narrative outcomes. Comparative evaluations have demonstrated CoDi's competitive narrative quality, highlighting its effectiveness in balancing structural control and flexible agent behaviors, thus underscoring its potential as a foundation for scalable interactive storytelling systems. The code is publicly available on Github.
Zitieren
APA
Kim, H., Yoo, T., & Cheong, Y. (2025). CoDi: A Director-Actor Framework for Goal-Driven Interactive Story Generation with LLMs. Proceedings of the AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment, 21(1), 70-80. https://doi.org/10.1609/aiide.v21i1.36811
Vancouver
Kim H, Yoo T, Cheong Y. CoDi: A Director-Actor Framework for Goal-Driven Interactive Story Generation with LLMs. Proceedings of the AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment. 2025;21(1):70-80. doi:10.1609/aiide.v21i1.36811
BibTeX
@article{Kim2025,
author = {Kim, Honggu and Yoo, Taewoo and Cheong, Yun-Gyung},
title = {CoDi: A Director-Actor Framework for Goal-Driven Interactive Story Generation with LLMs},
journal = {Proceedings of the AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment},
year = {2025},
volume = {21},
number = {1},
pages = {70--80},
doi = {10.1609/aiide.v21i1.36811}
}
EUVIMED