Large language models (LLMs) have ushered in a multitude of new language generation capabilities, bringing AI-guided storytelling much closer to reality. Authors can now meaningfully engage with AI-writing assitants to help during the story writing process. These same capabilities have also enabled players to experience more complex and diverse interactions in narrative-driven roleplaying games. Though due to the propensity of LLMs to confabulate facts, careful examination of their generated text is necessary. I begin by exploring LLMs as AI-writing assistants on the online collaborative writing platform Storium, where real authors query a model for suggested story continuations and then edit them. I follow up by observing players' experiences interacting with dialogue generated by GPT-4 within the widely-acclaimed role-playing game "Disco Elyisum: The Final Cut." Then, to reduce confabulation in generated text, I investigate techniques to boost factuality in LLMs by performing rank reductions on a subset of the matrices of a trained Transformer. Finally, looking to the future, I outline promising research directions for expanding AI-powered dialogue authoring tools for games.
Advisor: Mohit Iyyer