---
title: "SurveyForge: On the Outline Heuristics, Memory-Driven Generation, and Multi-dimensional Evaluation for Automated Survey Writing"
canonical_url: "https://www.modelscope.cn/papers/123886"
md_url: "https://www.modelscope.cn/papers/123886.md"
arxiv_id: 2503.04629
published: 2025-03-06
last_updated: 2025-03-06
authors:
  - "Xiangchao Yan"
  - "Shiyang Feng"
  - "Jiakang Yuan"
  - "Renqiu Xia"
  - "Bin Wang"
  - "Bo Zhang"
  - "Lei Bai"
model_name: SURVEYFORGE
model_developer: "上海人工智能实验室, 复旦大学, 上海交通大学"
domain:
  - "自然语言处理"
  - "机器学习"
  - "计算机科学"
type:
  - "自然语言处理"
  - "机器学习"
  - "计算机科学"
  - "Computation and Language (cs.CL)"
arxiv_url: "https://arxiv.org/abs/2503.04629"
pdf_url: "https://arxiv.org/pdf/2503.04629.pdf"
---

# SurveyForge: On the Outline Heuristics, Memory-Driven Generation, and Multi-dimensional Evaluation for Automated Survey Writing

> Survey paper plays a crucial role in scientific research, especially given the rapid growth of research publications. Recently, researchers have begun using LLMs to automate survey generation for better efficiency. However, the quality gap between…

「SurveyForge: On the Outline Heuristics, Memory-Driven Generation, and Multi-dimensional Evaluation for Automated Survey Writing」是 ModelScope 魔搭社区收录的论文，arXiv 2503.04629，作者为 Xiangchao Yan, Shiyang Feng, Jiakang Yuan et al.，发表于 2025-03-06，属于 自然语言处理、机器学习、计算机科学 领域。

- **ArXiv**: 2503.04629
- **Published**: 2025-03-06
- **Authors**: Xiangchao Yan, Shiyang Feng, Jiakang Yuan, Renqiu Xia, Bin Wang, Bo Zhang, Lei Bai
- **Model**: SURVEYFORGE
- **Developer**: 上海人工智能实验室, 复旦大学, 上海交通大学
- **Domain**: 自然语言处理, 机器学习, 计算机科学
- **ArXiv URL**: https://arxiv.org/abs/2503.04629
- **PDF**: https://arxiv.org/pdf/2503.04629.pdf

Source: https://www.modelscope.cn/papers/123886

---

> SURVEYFORGE：用启发式大纲生成和记忆驱动的内容创作提升自动综述写作的新标准

## 摘要

本文旨在解决AI生成的综述文章在结构逻辑和引用准确性方面与人工撰写的综述文章之间的质量差距。提出了一种名为SURVEYFORGE的自动化框架，该框架通过两个阶段来生成高质量的学术综述文章：大纲生成和内容生成。在大纲生成阶段，SURVEYFORGE利用启发式学习方法分析主题相关文献和现有综述的结构模式，生成语义丰富且组织良好的大纲。在内容生成阶段，基于记忆驱动的学者导航代理（SANA）检索高质量文献，并结合大型语言模型（LLM）生成各部分内容。此外，作者还构建了SurveyBench基准，用于多维度评估自动生成的综述文章，包括参考文献、大纲质量和内容质量。实验结果表明，SURVEYFORGE在多个维度上显著优于先前的方法，如AutoSurvey，不仅提高了生成内容的质量，还缩小了AI生成与人工撰写综述之间的差距。

## Abstract

Survey paper plays a crucial role in scientific research, especially given the rapid growth of research publications. Recently, researchers have begun using LLMs to automate survey generation for better efficiency. However, the quality gap between LLM-generated surveys and those written by human remains significant, particularly in terms of outline quality and citation accuracy. To close these gaps, we introduce SurveyForge, which first generates the outline by analyzing the logical structure of human-written outlines and referring to the retrieved domain-related articles. Subsequently, leveraging high-quality papers retrieved from memory by our scholar navigation agent, SurveyForge can automatically generate and refine the content of the generated article. Moreover, to achieve a comprehensive evaluation, we construct SurveyBench, which includes 100 human-written survey papers for win-rate comparison and assesses AI-generated survey papers across three dimensions: reference, outline, and content quality. Experiments demonstrate that SurveyForge can outperform previous works such as AutoSurvey.