Story2Proposal: A Scaffold for Structured Scientific Paper Writing

TL;DR

Story2Proposal is a multi-agent framework that enhances scientific paper writing by maintaining structural and visual consistency through contract-guided collaboration, outperforming existing methods in expert evaluations.

Contribution

It introduces a contract-governed multi-agent system that coordinates various agents to produce structured, visually aligned scientific manuscripts with improved consistency.

Findings

Achieved higher expert evaluation scores than baseline methods.

Improved structural consistency and visual alignment in generated papers.

Outperformed existing structured generation baselines in experiments.

Abstract

Generating scientific manuscripts requires maintaining alignment between narrative reasoning, experimental evidence, and visual artifacts across the document lifecycle. Existing language-model generation pipelines rely on unconstrained text synthesis with validation applied only after generation, often producing structural drift, missing figures or tables, and cross-section inconsistencies. We introduce Story2Proposal, a contract-governed multi-agent framework that converts a research story into a structured manuscript through coordinated agents operating under a persistent shared visual contract. The system organizes architect, writer, refiner, and renderer agents around a contract state that tracks section structure and registered visual elements, while evaluation agents supply feedback in a generate evaluate adapt loop that updates the contract during generation. Experiments on tasks derived from the Jericho research corpus show that Story2Proposal achieved an expert evaluation score of 6.145 versus 3.963 for DirectChat (+2.182) across GPT, Claude, Gemini, and Qwen backbones. Compared with the structured generation baseline Fars, Story2Proposal obtained an average score of 5.705 versus 5.197, indicating improved structural consistency and visual alignment.