Tag
This paper analyzes how Large Language Models (LLMs) use worldbuilding strategies in AI-generated creative storytelling, comparing them to human-authored fiction and finding that LLMs overproduce 'perceived space' while humans favor 'action space'.