Fictional Worldbuilding: Multi-Agent LLM Collaboration with Hierarchical Context Compression and Iterative Review
Quick Answer
This paper shows that The AutoWorldBuilder system utilizes multi-agent collaboration and hierarchical context compression to enhance fictional worldbuilding, achieving a 95.0% success rate across 20 tasks with LLMs like GPT-OSS 120B.
Quick Take
The AutoWorldBuilder system utilizes collaboration and hierarchical context compression to enhance fictional worldbuilding, achieving a 95.0% success rate across 20 tasks with LLMs like GPT-OSS 120B. It effectively reduces token usage by 90% and improves proposal pass rates from 42% to over 85% through iterative review and conflict detection mechanisms.
Key Points
- Achieved 95.0% success rate in worldbuilding tasks using GPT-OSS 120B.
- Reduced token usage by approximately 90% with a four-layer context compression.
- Improved proposal pass rates from 42% to over 85% with iterative review.
- Utilized a DAG-based hybrid batch scheduler for semantic locality in task grouping.
- Supported zero-code extension with differentiated temperature configuration in agents.
Paper Resources
📖 Reader Mode
~2 min readAbstract:Worldbuilding, the construction of coherent fictional worlds, is a foundational task in game design and literary creation. Large Language Models (LLMs) offer new possibilities for automated content generation, but their application to worldbuilding faces three challenges: context explosion that grows linearly with the building process, the tension between creative diversity and content consistency, and the absence of automated quality assurance. This paper presents AutoWorldBuilder, a multi-agent collaborative system that addresses these challenges through five integrated components: a structured concept network with conflict detection; a DAG-based hybrid batch scheduler that groups tasks by semantic locality; a four-layer context compression mechanism achieving approximately 90% token reduction; an iterative review system with specialized Auditor agents that improves proposal pass rates from 42% to over 85%; and a skill-driven agent architecture supporting zero-code extension with differentiated temperature configuration. Two experiments across 20 diverse worldbuilding tasks, using GPT-OSS 120B and DeepSeek v3.2 as LLM backends, demonstrate a 95.0% success rate. The system generated 56-103 self-consistent concepts per world in 18-31 minutes with zero-conflict delivery. The architectural patterns validated here, including layer-as-budget compression, semantic-locality scheduling, and separation of generation and review, transfer to the broader class of knowledge-intensive, multi-agent LLM applications.
| Comments: | 36 pages, 7 fig |
| Subjects: | Artificial Intelligence (cs.AI) |
| ACM classes: | I.2.11; I.2.4 |
| Cite as: | arXiv:2607.09403 [cs.AI] |
| (or arXiv:2607.09403v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2607.09403 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Jingbo Chen [view email]
[v1]
Fri, 10 Jul 2026 13:30:42 UTC (733 KB)
— Originally published at arxiv.org
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from arXiv cs.AI
See more →Adversarial Social Epistemology for Assemblies of Humans and Large Language Models
The paper introduces Adversarial Social Epistemology (ASE) to analyze how agents manipulate trust in public communications, highlighting mechanisms that undermine the reliability of testimony and inference. It critiques existing frameworks like epistemic bubbles and misinformation diffusion, proposing a new language for understanding trust breaches and auditing inferential chains in densely interactive environments involving humans and large language models.