PREFER framework: Guiding personalized game refinement with player experience
What happened
Researcher Qi Chai has published a paper proposing PREFER, a framework for evaluating how large language models judge game experiences and for tailoring game refinements to different player personas based on those judgments. The framework uses a set of experience dimensions adapted from established psychometric questionnaires, both to measure how the model evaluates a game and to guide adjustments for specific player personas. The paper reports that in experiments with five player personas, PREFER's post-play preference scores averaged 9.2% higher than the strongest comparison baseline. The result comes from the paper's own experimental report and has not yet been independently verified.
AI-written from the coverage · updated 2026-10-15
Coverage timeline
Follow the coverage to see the story from different sides.
- arXiv Game AIPREFER: From player experience to personalized game refinement
The PREFER framework uses a set of experience dimensions drawn from established psychometric questionnaires to evaluate how large language models (LLMs) judge game experiences, and to tailor game refinements for different player personas. In experiments with five player personas, PREFER's post-play preference scores averaged 9.2% higher than the strongest comparison baseline.