Back to Details

survey-generation

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

Survey Generation

综述论文生成

Generate complete academic survey papers with structured outline, RAG-based writing, and citation validation.
借助结构化大纲、基于RAG的写作及引用验证功能，生成完整的学术综述论文。

Input

输入

  • $0
    — Survey topic or research area
  • $0
    — 综述主题或研究领域

Scripts

脚本

Literature search

文献检索

bash
python ~/.claude/skills/deep-research/scripts/search_semantic_scholar.py \
  --query "relevant search query" --max-results 50
bash
python ~/.claude/skills/deep-research/scripts/search_semantic_scholar.py \
  --query "relevant search query" --max-results 50

References

参考资源

  • Survey prompts (outline, writing, citation, coherence): 
    ~/.claude/skills/survey-generation/references/survey-prompts.md
  • 综述提示词（大纲、写作、引用、连贯性）：
    ~/.claude/skills/survey-generation/references/survey-prompts.md

Workflow (from AutoSurvey)

工作流（源自AutoSurvey）

Step 1: Collect Papers

步骤1：收集文献

  1. Search Semantic Scholar / arXiv for papers on the topic
  2. Collect 50-200 relevant papers with titles and abstracts
  3. Filter by relevance and citation count
  1. 在Semantic Scholar / arXiv上检索该主题的文献
  2. 收集50-200篇相关文献的标题及摘要
  3. 按相关性和引用量筛选文献

Step 2: Generate Outline (Multi-LLM Parallel)

步骤2：生成大纲（多LLM并行）

  1. Generate N rough outlines independently (parallel)
  2. Merge outlines into a single comprehensive outline
  3. Expand each section into subsections
  4. Edit final outline to remove redundancies
  1. 独立生成N份初步大纲（并行处理）
  2. 将多份大纲合并为一份综合大纲
  3. 将每个章节扩展为小节
  4. 编辑最终大纲，移除冗余内容

Step 3: Write Subsections (RAG-Based)

步骤3：撰写小节（基于RAG）

For each subsection:
  1. Retrieve relevant papers for the subsection topic
  2. Generate content with inline citations 
    [paper_title]
  3. Enforce minimum word count per subsection
  4. Only cite papers from the provided list
针对每个小节：
  1. 检索与该小节主题相关的文献
  2. 生成带有内嵌引用 
    [paper_title]
    的内容
  3. 确保每个小节达到最低字数要求
  4. 仅引用提供的文献列表中的文献

Step 4: Validate Citations

步骤4：验证引用

For each subsection:
  1. Check that cited paper titles are correct
  2. Verify citations support the claims made
  3. Remove or correct unsupported citations
  4. Use NLI (Natural Language Inference) for claim-source faithfulness
针对每个小节：
  1. 检查引用的文献标题是否正确
  2. 验证引用是否支撑文中观点
  3. 移除或修正无支撑的引用
  4. 使用自然语言推理（NLI）确保观点与来源的一致性

Step 5: Enhance Local Coherence

步骤5：增强局部连贯性

For each subsection:
  1. Read previous and following subsections
  2. Refine transitions and flow
  3. Preserve core content and citations
  4. Ensure smooth reading experience
针对每个小节：
  1. 阅读前后小节内容
  2. 优化过渡与行文流畅度
  3. 保留核心内容与引用
  4. 确保阅读体验流畅

Step 6: Convert Citations to BibTeX

步骤6：将引用转换为BibTeX格式

  1. Replace 
    [paper_title]
    with 
    \cite{key}
  2. Generate BibTeX entries for all cited papers
  3. Validate all citation keys exist in .bib file
  1. 将 
    [paper_title]
    替换为 
    \cite{key}
  2. 为所有引用的文献生成BibTeX条目
  3. 验证所有引用标识均存在于.bib文件中

Output Structure

输出结构

survey/
├── main.tex          # Complete survey paper
├── references.bib    # All citations
├── outline.json      # Survey outline
└── sections/         # Individual section files
survey/
├── main.tex          # 完整的综述论文
├── references.bib    # 所有引用文献
├── outline.json      # 综述大纲
└── sections/         # 各章节独立文件

Rules

规则

  • Only cite papers from the collected paper list — never hallucinate citations
  • Each subsection must meet minimum word count
  • No duplicate subsections across sections
  • Citation validation is mandatory before final output
  • Local coherence enhancement must preserve all citations
  • The survey should be comprehensive and logically organized
  • 仅引用收集到的文献列表中的文献——禁止虚构引用
  • 每个小节必须达到最低字数要求
  • 各章节间不得存在重复小节
  • 最终输出前必须完成引用验证
  • 局部连贯性增强操作必须保留所有引用
  • 综述内容需全面且逻辑清晰

Related Skills

相关技能

  • Upstream: deep-research, literature-search, literature-review
  • See also: related-work-writing
  • 上游技能：深度研究、文献检索、文献综述
  • 相关技能：研究背景撰写