The paper ‘Skill Check: Some Considerations on the Evaluation of Gamemastering Models for Role-playing Games’ discusses the challenges in modelling Game Masters (GMs) for role-playing games from an Interactive Storytelling and Natural Language Processing perspective. The authors propose three test categories to evaluate such dialogue systems and use them to test ChatGPT, Bard, and OpenAssistant as out-of-the-box GMs. The goal is to model the diverse skills needed by a GM to play Role-Playing Games. This research is a step towards creating an explainable, grounded, and controllable model.
Publication date: 24 Sep 2023
Project Page: arXiv:2309.13702v1 [cs.CL] Paper: https://arxiv.org/pdf/2309.13702
