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'.
This paper presents AutoWorldBuilder, a multi-agent LLM system for automated fictional worldbuilding that addresses context explosion, creative diversity, and quality assurance through hierarchical context compression, DAG-based scheduling, and iterative review, achieving 95% success rate and generating 56–103 self-consistent concepts per world.