BookWorld: From Novels to Interactive Agent Societies for Creative Story Generation
This work addresses the underexplored area of simulating established fictional worlds for applications like story generation and interactive games, offering a novel approach to extend beloved fictional works.
The paper tackles the problem of simulating established fictional worlds and characters for creative story generation by introducing BookWorld, a system for constructing and simulating book-based multi-agent societies, which achieves a win rate of 75.36% in generating high-quality stories while maintaining fidelity to source books.
Recent advances in large language models (LLMs) have enabled social simulation through multi-agent systems. Prior efforts focus on agent societies created from scratch, assigning agents with newly defined personas. However, simulating established fictional worlds and characters remain largely underexplored, despite its significant practical value. In this paper, we introduce BookWorld, a comprehensive system for constructing and simulating book-based multi-agent societies. BookWorld's design covers comprehensive real-world intricacies, including diverse and dynamic characters, fictional worldviews, geographical constraints and changes, e.t.c. BookWorld enables diverse applications including story generation, interactive games and social simulation, offering novel ways to extend and explore beloved fictional works. Through extensive experiments, we demonstrate that BookWorld generates creative, high-quality stories while maintaining fidelity to the source books, surpassing previous methods with a win rate of 75.36%. The code of this paper can be found at the project page: https://bookworld2025.github.io/.