Spark-to-Paper: End-to-End Research Paper Generation as a Composable Skill

2026-08-12

AI Summary

This work addresses the challenge of transforming research ideas into complete academic papers—a process requiring coordinated literature retrieval, experimental design, evidence alignment, and long-form coherence, which conventional generation methods struggle to support. The authors propose a lightweight, composable workflow architecture embedded within a coding assistant, integrating thirteen modular skills to enable end-to-end paper generation. Their approach decouples model-based judgments from deterministic operations and introduces a novel separation between experimental planning and reporting. By incorporating evidence-driven claim revision, self-critique mechanisms, and procedural vector graphic generation, the system effectively mitigates failure modes such as self-contradictory loops. Evaluated across eight controlled tasks, the framework achieves 99.5% citation validity, 96.4% editable figure fidelity, reduces hallucination rates from 14% to 8% (corresponding to a 92% detection rate), attains 74% accuracy under adversarial review, and produces each paper in an average of 3.2 hours at a cost of $8.10.

Abstract

Turning a research idea into a complete paper requires more than text generation: the system must retrieve literature, design and execute experiments, revise claims according to evidence, produce publication-ready figures, and maintain consistency across a long generation process. We present Spark-to-Paper, an end-to-end research paper generation system implemented as thirteen composable skills inside an existing coding assistant, without requiring a separate agent platform or orchestration service. Spark-to-Paper separates model-based judgment from deterministic operations that can be directly executed and checked. It further separates experiment planning from reporting, so that required evidence is specified before results are observed and manuscript claims are revised according to measured outcomes. To improve reliability over long research trajectories, the system combines deterministic integrity checks with self-critique and bounds a failure mode we call the Self-Refutation Loop, in which repeated experiments continue to reject the original research objective. Spark-to-Paper also produces editable vector figures through programmatic plotting for experimental results and code-based reconstruction for generated method diagrams. Across eight controlled research topics, Spark-to-Paper achieves 99.5% citation validity and 96.4% figure editability. A controlled ablation increases fabrication detection from 14% for a single-pass draft to 92% with the full integrity and review stack, while adversarial review achieves 74% precision. The full system uses 11.9M tokens, costs $8.1 per manuscript, and requires 3.2 hours on average. These results show that end-to-end research paper generation can be implemented as a lightweight, composable workflow inside existing coding assistants while keeping experimental evidence central to how claims are accepted, revised, or abandoned.

Problem

• research paper generation
• end-to-end automation
• experimental evidence
• claim consistency
• scientific integrity

Innovation

• composable skills
• end-to-end paper generation
• evidence-driven revision
• self-refutation loop mitigation
• programmatic figure generation