Plot-Logic-Pipeline
Systematically deconstruct scientific papers by following the figure-discussion logical backbone.
When to Use
- - Analyzing a research paper's argument structure
- Reviewing manuscripts before submission
- Understanding how figures support claims in technical papers
- Mapping evidence sources (literature vs. new measurements)
- Identifying logical gaps or unsupported claims
Core Principle
Figures are the bare bones of a paper's logic flow. Each figure corresponds to a discussion that either:
- - Sets up the next key finding (preparation)
- States the key finding (conclusion)
Complete understanding requires analyzing every figure-discussion pair and tracking evidence sources.
Analysis Framework
Step 1: Figure Inventory
Create a complete inventory of all figures in the paper:
CODEBLOCK0
Step 2: Figure-Discussion Mapping
For each figure, identify its corresponding discussion section and analyze:
CODEBLOCK1
Step 3: Logic Flow Reconstruction
Map how figures build upon each other:
CODEBLOCK2
Step 4: Evidence Assessment
Evaluate the strength of the paper's argument:
- - Are all major claims supported by figures?
- Are evidence sources properly attributed?
- Are there logical gaps between figures?
- Do setup discussions adequately prepare for key findings?
Evidence Classification
Previous Study Support
- - Direct citation: Specific reference supporting the claim
- Literature consensus: Multiple citations building consensus
- Comparative reference: Contrasting with previous work
This Paper's Contributions
- - New experimental data: Novel measurements with method specified
- Novel calculations: Computational work or modeling
- Reanalysis: New interpretation of existing data
Combined Evidence
- - Validation: New data confirms previous studies
- Extension: New data builds upon previous work
- Contradiction: New data challenges previous findings
Analysis Templates
See TEMPLATES.md for detailed templates including:
- - Basic figure-discussion analysis
- Complete paper analysis workflow
- Materials science specific templates
- Quality assurance checklist
Quality Checks
Before concluding analysis:
- - ✅ All figures mapped to discussions
- ✅ Evidence sources identified for major claims
- ✅ Logic flow clearly traced from introduction to conclusion
- ✅ Setup vs. statement discussions distinguished
- ✅ Contradictions or gaps noted and flagged
Common Pitfalls
- - Skipping "obvious" figures: Even simple schematics contribute to logic flow
- Missing evidence attribution: Always identify if claims come from citations or new work
- Ignoring setup discussions: These are crucial for understanding logical progression
- Overlooking figure details: Axis labels, error bars, and annotations often contain key information
- Conflating correlation with causation: Note when figures show correlation vs. when claims assert causation
Rules
- 1. Every figure gets analyzed — no skipping, even if it seems straightforward
- Always classify evidence — distinguish previous work from new contributions
- Trace the logic chain — show how each figure builds on the previous one
- Flag gaps honestly — note missing evidence or weak logical connections
- Separate observation from interpretation — what the figure shows vs. what the authors claim
图表-逻辑-分析流程
通过遵循图表-讨论逻辑主线,系统性地解构科学论文。
适用场景
- - 分析研究论文的论证结构
- 投稿前审阅稿件
- 理解技术论文中图表如何支撑论点
- 映射证据来源(文献 vs. 新测量数据)
- 识别逻辑漏洞或缺乏支撑的论断
核心原则
图表是论文逻辑流程的骨架。 每个图表对应一段讨论,其作用要么是:
- - 铺垫下一个关键发现(准备)
- 陈述关键发现(结论)
完整理解需要分析每一组图表-讨论对,并追踪证据来源。
分析框架
第一步:图表清单
创建论文中所有图表的完整清单:
图1:[简要描述]
图2:[简要描述]
...
图N:[简要描述]
第二步:图表-讨论映射
针对每个图表,识别其对应的讨论部分并进行分析:
图X:[描述]
├── 位置:[讨论该图所在的章节/页码]
├── 讨论类型:[准备 / 陈述]
├── 核心论断:[关键发现或要点]
└── 证据来源:
├── 先前研究:[若由文献支持,列出引用]
├── 本文:[若为新测量/计算,注明分析方法]
└── 支撑程度:[强 / 部分 / 矛盾 / 缺失]
第三步:逻辑流程重构
映射图表之间的递进关系:
论文逻辑流程:
图1 → 图2 → 图3 → ... → 结论
↓ ↓ ↓
[准备] [关键发现1] [关键发现2]
第四步:证据评估
评估论文论证的强度:
- - 所有主要论断是否都有图表支撑?
- 证据来源是否恰当归属?
- 图表之间是否存在逻辑断层?
- 准备性讨论是否充分铺垫了关键发现?
证据分类
先前研究支撑
- - 直接引用:支持论断的具体参考文献
- 文献共识:多篇引用构建的共识
- 对比参考:与先前工作的对比
本文贡献
- - 新实验数据:注明方法的新测量结果
- 新计算:计算工作或建模
- 再分析:对现有数据的新解读
综合证据
- - 验证:新数据证实先前研究
- 拓展:新数据建立在先前工作基础上
- 矛盾:新数据挑战先前发现
分析模板
详见 TEMPLATES.md 中的详细模板,包括:
- - 基础图表-讨论分析
- 完整论文分析流程
- 材料科学专用模板
- 质量保证检查清单
质量检查
在结束分析前:
- - ✅ 所有图表均已映射到讨论
- ✅ 主要论断的证据来源已识别
- ✅ 从引言到结论的逻辑流程已清晰追踪
- ✅ 已区分准备性讨论与陈述性讨论
- ✅ 矛盾或断层已记录并标记
常见陷阱
- - 跳过显而易见的图表:即使是简单的示意图也对逻辑流程有贡献
- 遗漏证据归属:始终识别论断来自引用还是新工作
- 忽视准备性讨论:这对理解逻辑推进至关重要
- 忽略图表细节:坐标轴标签、误差线和注释常包含关键信息
- 混淆相关性与因果性:注意图表展示的是相关性还是论断声称的因果关系
规则
- 1. 每个图表都要分析——即使看起来很简单,也不可跳过
- 始终对证据进行分类——区分先前工作与新贡献
- 追踪逻辑链条——展示每个图表如何建立在之前图表的基础上
- 诚实标记断层——记录缺失的证据或薄弱的逻辑连接
- 区分观察与解读——图表显示的内容 vs. 作者声称的内容