Adnan KhanAdnanProduct & AI
San Francisco Bay Area
Get in touch

Source note2 min read

From Personas to Plot: Character-Grounded Multi-Agent Story Generation for Long-Form Narratives

Source

arXiv

arxiv.org/abs/2607.00918v1

Read the source

Source

Plain-Language Brief

Researchers have developed MAGNET, a framework that generates consistent and coherent long-form narratives by using a unified approach of multi-agent goal-driven storytelling and verification. The system was tested against single-model prompting and IBSEN, with results showing significant reduction in annotations and hallucinations compared to the baselines.

What the Document Says

  • Researchers have introduced MAGNET, a framework for long-form narrative generation and verification.
  • MAGNET generates stories using persona-grounded character agents that propose actions based on shared world states and evolving story goals.
  • ATLAS is a graph-based pipeline used to compare scene-level world representations across a generated story to detect hallucinations.
  • The researchers evaluated MAGNET using an LLM editor, pairwise rubric scoring, and ATLAS.
  • The evaluation showed that MAGNET produces coherent narratives compared to single-model prompting and IBSEN.
  • At 100 pages, MAGNET reduced annotations by 41% and hallucinations by 50% compared to the single model baseline.
  • Compared to IBSEN, MAGNET reduced annotations by 34% and hallucinations by 45%.
  • Pairwise rubric evaluation showed similar results.
  • The results suggest that long-form narratives can emerge from explicit world-state tracking and goal-driven multi-agent generation.
  • This provides a foundation for controllable and structurally coherent long-form narrative generation.

Key Concepts Mentioned

  • long-form narratives
  • narrative consistency
  • coherent plot lines
  • multi-agent story generation
  • world state tracking
  • goal-driven multi-agent generation
  • explicit world-state tracking

Claims Made by the Document

  • MAGNET produces coherent narratives compared to single-model prompting and IBSEN.
  • MAGNET reduces annotations and hallucinations significantly compared to the baselines.

Evidence Used

  • The evaluation results provided by the researchers

Methods / Approach

UNKNOWN

Limitations Stated by the Authors

UNKNOWN

Limitations Noted by the Agent

Agent interpretation — not from the source.

UNKNOWN

Open Questions

UNKNOWN

Notes for Future Q&A

May include interpretation. Labeled where it does.

UNKNOWN

Raw Abstract / Excerpt

Although large language models (LLMs) have demonstrated impressive creative fiction generation, they struggle to maintain narrative consistency and coherent plot lines in long-form stories. In this work, we introduce a unified framework for long-form narrative generation and verification. MAGNET, a multi-agent goal-driven narrative engine for storytelling, generates stories with persona-grounded character agents that propose actions based on a shared world state and evolving story goals, while ATLAS is a graph-based pipeline that compares scene-level world representations across a generated story to detect hallucinations. By evaluating MAGNET using an LLM editor, pairwise rubric scoring, and ATLAS, we show that our framework produces coherent narratives compared to single-model prompting and IBSEN. At 100 pages, MAGNET reduced annotations and hallucinations by 41 and 50%, respectively, compared to the single model baseline and by 34 and 45%, respectively, compared to IBSEN, with pairwise rubric evaluation showing similar results. These results suggest that long-form narratives can emerge from explicit world-state tracking and goal-driven multi-agent generation, providing a foundation for controllable and structurally coherent long-form narrative generation.