# Spark-to-Paper：将端到端研究论文生成实现为可组合技能

- 来源：HuggingFace Daily Papers（社区热门论文）
- 发布时间：2026-08-12 08:00
- AIHOT 分数：54
- AIHOT 链接：https://aihot.virxact.com/items/cmsqvi08g06ilroz2z8k46oqa
- 原文链接：https://arxiv.org/abs/2608.11924

## AI 摘要

Spark-to-Paper 在现有编码助手中以十三个可组合技能实现端到端研究论文生成，无需独立智能体平台或编排服务。系统将模型判断与可执行确定性操作分离，并通过完整性检查与自我批判将“自我反驳循环”失败模式限制在可控范围。在八个受控研究主题上，论文引用有效性达 99.5%，图表可编辑性达 96.4%；完整系统每篇论文耗 11.9M tokens、成本 $8.1、平均耗时 3.2 小时。

## 正文

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.