Zero-Shot Reasoning: Personalized Content Generation Without the Cold Start Problem

Procedural content generation uses algorithmic techniques to create large amounts of new content for games at much lower production costs. To improve its quality, in newer approaches, procedural content generation utilizes machine learning. However, these methods usually require expensive collection of large amounts of data, as well as the development and training of fairly complex learning models, which can be both extremely time-consuming and expensive. The core of our research is to explore whether we can lower the barrier to the use of personalized procedural content generation through a more practical and generalizable approach with large language models. Matching game content to player preferences benefits both players, by enhancing enjoyment, and developers, who rely on player satisfaction for monetization. Therefore, this paper introduces a new method for personalization by using large language models to suggest levels based on ongoing gameplay data from each player. We compared the levels generated using our approach with levels generated with more traditional procedural generation techniques. Our easily reproducible method has proven viable in a production setting and outperformed levels generated by traditional methods in two aspects-the player's rating of levels and the probability that a player will not quit the game mid-level.