Agents' Room: Narrative Generation through Multi-step Collaboration

TL;DR

Agents' Room is a collaborative framework that decomposes narrative generation into subtasks handled by specialized agents, improving story quality over baseline models through structured collaboration and evaluation.

Contribution

It introduces a novel multi-agent collaboration approach for narrative generation, inspired by narrative theory, and provides a new dataset and evaluation framework for long story assessment.

Findings

Generated stories preferred by experts over baselines

Collaboration improves narrative coherence and quality

New dataset and evaluation framework for long narratives

Abstract

Writing compelling fiction is a multifaceted process combining elements such as crafting a plot, developing interesting characters, and using evocative language. While large language models (LLMs) show promise for story writing, they currently rely heavily on intricate prompting, which limits their use. We propose Agents' Room, a generation framework inspired by narrative theory, that decomposes narrative writing into subtasks tackled by specialized agents. To illustrate our method, we introduce Tell Me A Story, a high-quality dataset of complex writing prompts and human-written stories, and a novel evaluation framework designed specifically for assessing long narratives. We show that Agents' Room generates stories that are preferred by expert evaluators over those produced by baseline systems by leveraging collaboration and specialization to decompose the complex story writing task into tractable components. We provide extensive analysis with automated and human-based metrics of the generated output.