StoryBox: Collaborative Multi-Agent Simulation for Hybrid Bottom-Up Long-Form Story Generation Using Large Language Models

2025-10-13

AI Summary

Existing long-form story generation models struggle to maintain both coherence and consistency at scales exceeding 10,000 characters; conventional top-down approaches suffer from structural rigidity, hindering natural plot evolution and character development. Method: We propose a hybrid bottom-up generation paradigm leveraging a large language model–based multi-agent system. Agents collaboratively reason and interact with a dynamic sandbox environment to emergently drive character behavior and narrative events, enabling organic plot progression and character arc formation. Contribution/Results: Our approach eliminates reliance on predefined narrative structures and supports end-to-end generation of coherent, 10,000+-character stories. Quantitative and human evaluations demonstrate state-of-the-art performance across coherence, consistency, and plot richness metrics.

Technology Category

• Humans and AI: Game Design — Procedural Content Generation & Storytelling
• Natural Language Processing: Generation
• Multiagent Systems: Agent-Based Simulation and Emergent Behavior

Application Category

• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications

Abstract

Human writers often begin their stories with an overarching mental scene, where they envision the interactions between characters and their environment. Inspired by this creative process, we propose a novel approach to long-form story generation, termed hybrid bottom-up long-form story generation, using multi-agent simulations. In our method, agents interact within a dynamic sandbox environment, where their behaviors and interactions with one another and the environment generate emergent events. These events form the foundation for the story, enabling organic character development and plot progression. Unlike traditional top-down approaches that impose rigid structures, our hybrid bottom-up approach allows for the natural unfolding of events, fostering more spontaneous and engaging storytelling. The system is capable of generating stories exceeding 10,000 words while maintaining coherence and consistency, addressing some of the key challenges faced by current story generation models. We achieve state-of-the-art performance across several metrics. This approach offers a scalable and innovative solution for creating dynamic, immersive long-form stories that evolve organically from agent-driven interactions.

Problem

• Generating coherent long-form stories exceeding 10,000 words
• Enabling organic character development and plot progression
• Creating dynamic stories through multi-agent interactions

Innovation

• Multi-agent simulation generates emergent story events
• Hybrid bottom-up approach enables organic plot progression
• Dynamic sandbox environment fosters spontaneous character interactions