Back to Details

figure-rhetoric

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

Figure Rhetoric Audit

Figure修辞审计

Pipeline position: Phase 1b (content audit). Runs in parallel with manuscript-review. Requires compiled PDF. No prior dependencies. See 
/manuscript-pipeline
for full execution order.
流水线位置: 阶段1b（内容审核）。与manuscript-review并行运行。需要已编译的PDF文件。无前置依赖。查看
/manuscript-pipeline
获取完整执行顺序。

Purpose

目的

Evaluate every figure in a manuscript as a rhetorical act — a visual argument that must land with the reader. Each figure exists to communicate a specific claim. This skill audits whether the figure actually achieves that communication, or whether it undermines, obscures, or contradicts the author's intent.
A figure that is technically correct but rhetorically ineffective is a wasted opportunity. Reviewers form judgments from figures before reading the methodology. A figure that fails to show what the text claims creates doubt even when the underlying data supports the claim.
将手稿中的每个Figure视为一种修辞行为——一种必须让读者理解的视觉论证。每个Figure都旨在传达特定论点。本技能审核该Figure是否真正实现了这种传达，或者是否削弱、模糊或违背了作者的意图。
技术上正确但修辞效果不佳的Figure是一种浪费的机会。审稿人在阅读方法论之前就会通过Figures形成判断。如果Figure未能展示文本所声称的内容，即使基础数据支持该论点，也会引发质疑。

Relationship to Other Skills

与其他技能的关系

ConcernThis skill (figure-rhetoric)manuscript-reviewmanuscript-typography
Chart type selectionIs this the right chart for this claim?N/AN/A
Visual emphasisDoes the figure draw attention to the right thing?N/AN/A
Prose-figure alignmentDoes a reader SEE what the text SAYS?Does the text match the figure? (§24)N/A
Data selectionShould different data be plotted?N/AN/A
Axis designDo axes help or hide the story?Axis labels present? (§12)Font consistency
Figure qualityN/AResolution, colorblind, chartjunk (§12)Backgrounds, framing
Figure renderingN/ALegibility at print scale (§23)Caption formatting
ProvenanceN/AN/A (→ manuscript-provenance)N/A
Boundary: manuscript-review §24 checks "does the prose match the figure?" This skill checks "does the figure communicate what the prose needs it to?" §24 catches factual mismatches (text says 14.3%, figure shows 13.8%). This skill catches rhetorical failures (text says "dramatic improvement," figure shows bars that look identical because the y-axis starts at 0).
关注点本技能（figure-rhetoric）manuscript-reviewmanuscript-typography
图表类型选择该图表是否适合此论点？不涉及不涉及
视觉强调Figure是否将注意力吸引到正确的内容上？不涉及不涉及
文本-Figure一致性读者是否能看到文本所描述的内容？文本与Figure是否匹配？（§24）不涉及
数据选择是否应绘制不同的数据？不涉及不涉及
坐标轴设计坐标轴是否有助于呈现故事还是隐藏故事？是否有坐标轴标签？（§12）字体一致性
Figure质量不涉及分辨率、色盲友好性、图表冗余信息（§12）背景、边框
Figure渲染不涉及打印尺度下的易读性（§23）图表说明格式
来源不涉及不涉及（→ manuscript-provenance）不涉及
边界说明: manuscript-review §24检查“文本与Figure是否匹配？”。本技能检查“Figure是否能传达文本所需的内容？”。§24捕捉事实性不匹配（文本称14.3%，Figure显示13.8%）。本技能捕捉修辞失效（文本称“显著改进”，但Figure显示的条形图因Y轴从0开始看起来几乎相同）。

Workflow

工作流程

1. Ingest

1. 导入

CRITICAL: This skill requires visual inspection. LaTeX source alone is insufficient. The entire point of this audit is what the reader sees.
Obtain the rendered figures by one of:
  1. Compiled PDF (preferred) — use the Read tool on the PDF file to inspect each page containing a figure at actual rendered size
  2. Individual figure files — use the Read tool on each 
    .pdf
    , 
    .png
    , 
    .jpg
    figure file referenced by 
    \includegraphics
  3. If neither is available — ask the user to compile the PDF or provide figure files. Do NOT proceed with source-only analysis. A rhetoric audit without seeing the figures is meaningless.
For each figure:
  1. Visually inspect the figure. Read the figure file or the PDF page containing it. Before reading any prose, record what the figure communicates at first glance — the immediate visual takeaway. What pattern, trend, comparison, or relationship does a naive reader see?
  2. Extract the claim context. Read the 2-3 paragraphs surrounding the first 
    \ref{fig:X}
    reference. Identify the specific claim the figure is supposed to support. Write it down as a one-sentence assertion.
  3. Read the caption. Does the caption tell the reader what to see, or does it just describe the axes?
  4. Compare. Does the visual takeaway (step 1) match the claimed assertion (step 2)? The figure must be evaluated through the reader's eyes, not the author's intent.
关键提示：本技能需要视觉检查。仅LaTeX源码是不够的。本次审计的核心是读者看到的内容。
通过以下方式获取渲染后的Figures：
  1. 已编译PDF（首选）——使用读取工具查看PDF文件中包含Figure的每一页的实际渲染尺寸
  2. 单个Figure文件——使用读取工具查看
    \includegraphics
    引用的每个
    .pdf
    、
    .png
    、
    .jpg
    格式的Figure文件
  3. 若以上均不可用——请用户编译PDF或提供Figure文件。不要仅基于源码进行分析。不查看Figures的修辞审计毫无意义。
针对每个Figure：
  1. 视觉检查Figure。读取Figure文件或包含它的PDF页面。在阅读任何文本之前，记录Figure在第一眼传达的信息——即时视觉印象。一个普通读者能看到什么模式、趋势、对比或关系？
  2. 提取论点上下文。阅读首次引用
    \ref{fig:X}
    前后的2-3个段落。确定该Figure应支持的具体论点。将其写成一句断言。
  3. 阅读图表说明。图表说明是告诉读者该看什么，还是仅描述坐标轴？
  4. 对比。视觉印象（步骤1）是否与声称的断言（步骤2）匹配？必须从读者的视角而非作者的意图来评估Figure。

2. Per-Figure Analysis

2. 单Figure分析

For each figure, evaluate across 8 dimensions:
Dimension 1 — Claim-Figure Alignment
The central question: does a reader who looks at this figure see the claim the text makes?
  • Identify the prose claim (e.g., "Method A converges faster than B")
  • Identify the visual impression (e.g., "Two nearly identical curves")
  • If these differ: the figure fails its rhetorical purpose regardless of whether the data technically supports the claim
Failure modes:
  • Invisible difference: The text claims a meaningful difference but the figure's scale, aspect ratio, or data density makes the difference imperceptible. The data is there; the visual isn't.
  • Wrong emphasis: The figure shows many things, the text discusses one. The reader doesn't know where to look.
  • Contradictory impression: The visual impression actively suggests the opposite of the claim (e.g., "convergence" but the curve is still trending; "improvement" but the bars look equal).
  • Unstated context: The figure requires domain knowledge to interpret that the surrounding text doesn't provide.
Dimension 2 — Chart Type Appropriateness
Is this the right type of visualization for this claim?
Claim typeEffective chartIneffective chart
Comparison across categoriesGrouped bar, dot plotPie chart, stacked bar (hard to compare)
Trend over time/sequenceLine plotBar chart (obscures continuity)
DistributionHistogram, violin, box plotBar chart of means (hides variance)
Correlation / relationshipScatter plotTable of paired values
Composition / proportionStacked area, stacked barMultiple pie charts
Difference / improvementDifference plot, waterfallSide-by-side bars at full scale
RankingHorizontal bar (sorted)Vertical bar (unsorted)
SpatialHeatmap, contourTable of values
Part-to-wholeTreemap, stacked barGrouped bar
Flow / processSankey, alluvialStatic diagram
Confusion / classificationConfusion matrix (heatmap)Table of numbers
Ablation contributionsWaterfall chartTable or grouped bar
Flag when a more effective chart type exists for the claim being made.
Dimension 3 — Axis and Scale Design
Axes can make or break the visual argument.
  • Y-axis origin: Starting at 0 when differences are small makes them invisible. Starting above 0 when showing absolute quantities is misleading. The choice must serve the claim:
    • Showing "our method is better" → zoom in on the relevant range
    • Showing "the effect is small relative to the total" → start at 0
    • Always: clearly label the axis range; broken axes with 
      //
      if needed
  • Axis scale: Linear vs logarithmic. Log scale appropriate when data spans orders of magnitude or when relative differences matter more than absolute. Flag linear scale with data spanning 3+ orders of magnitude.
  • Axis range: Does the range include all relevant data? Does it extend far beyond the data (wasting space)? Is the range chosen to exaggerate or minimize visual differences?
  • Aspect ratio manipulation: A very wide or very narrow plot can exaggerate or flatten trends. The slope of a trend line should be perceptible but not exaggerated.
  • Dual axes: Almost always confusing. Two different y-axes invite incorrect visual comparisons. Prefer: two separate panels with aligned x-axes.
Dimension 4 — Visual Hierarchy and Emphasis
Does the figure guide the reader's eye to the right element?
  • Primary element: The data series, region, or comparison that the text discusses should be visually dominant (thicker line, bolder color, larger markers, foreground position).
  • Secondary elements: Context, baselines, and reference data should be visually recessive (thinner lines, gray, smaller markers, background).
  • Annotations: If the text references a specific point, region, or threshold, it should be annotated in the figure (arrow, callout, shaded region, dashed reference line). The reader should not have to decode coordinates to find what the text describes.
  • Cognitive load: Count the number of distinct visual elements the reader must track. More than 5-7 series/groups in one plot exceeds working memory. Split into panels or highlight the comparison of interest.
Failure modes:
  • All lines/bars have equal visual weight — reader doesn't know what's important
  • The "our method" line is the same width and style as 10 baseline lines
  • Text says "note the divergence at epoch 100" but nothing in the figure marks epoch 100
  • Legend has 15 entries — reader must constantly reference it
Dimension 5 — Data Density and Simplification
Is the figure showing the right amount of data for its claim?
  • Overloaded: Too many series, categories, or data points for a single plot. The claim is about the relationship between 2 methods but the plot shows 12. Simplify: show the comparison of interest prominently; relegate others to supplementary material or a secondary panel.
  • Underloaded: The plot shows so little data that the claim isn't convincing. A single point where a trend is claimed. Two bars where a distribution is relevant. Three epochs where convergence behavior matters.
  • Summarized when raw matters: Showing only means when the variance is the story (or when it would undermine the story — flag both). Confidence intervals, error bars, or violin plots for stochastic results.
  • Raw when summary matters: Individual data points where the aggregate pattern is the claim. A scatter plot of 10,000 points where a density plot or binned heatmap would show the structure.
Dimension 6 — Caption as Interpretation Guide
The caption should tell the reader what to see, not just what the axes are.
Levels of caption quality:
  1. Descriptive only (weak): "Training loss over epochs for five methods." The reader must figure out the takeaway.
  2. Directive (adequate): "Training loss over epochs. Method A (red) converges in ~50 epochs while baselines require 150+." Points the reader to the claim.
  3. Interpretive (strong): "Training loss over epochs. Method A (red) converges 3x faster than the nearest baseline (blue), supporting the claim that architectural change X reduces optimization difficulty." Connects the visual to the argument.
For each caption, identify its level and recommend upgrading if Level 1. Level 2 is the minimum for effective communication. Level 3 is ideal for key figures supporting core claims.
Dimension 7 — Perceptual Accuracy
Does the visual encoding accurately represent the underlying data?
  • Area encoding for quantities: If using bar width, bubble size, or area to encode values, the mapping must be proportional to area (not radius or diameter). A value 2x larger should have 2x the area, not 2x the radius (which is 4x the area).
  • 3D effects: 3D bar charts, 3D pie charts, perspective effects — these distort perception of values through foreshortening. Flag any 3D visualization of 2D data.
  • Color scale linearity: Sequential color maps (viridis, plasma) have perceptually uniform steps. Rainbow/jet color maps have perceptual discontinuities that create artificial boundaries in continuous data. Flag rainbow/jet for continuous data.
  • Truncated axes without indication: Y-axis not starting at 0 without a visible break (
    //
    notation) can mislead readers about relative magnitudes.
  • Aspect ratio distortion: Pie charts not circular. Bar widths inconsistent. Maps with incorrect projections (rare in ML papers but common in spatial analysis).
Dimension 8 — Redundancy and Narrative Arc
Do the figures as a set tell a coherent story?
  • Redundant figures: Two figures showing essentially the same information in different forms. Unless each adds distinct insight, merge or choose the more effective one.
  • Missing figures: Is there a key claim in the text that has no visual support but would benefit from one? A figure that isn't there is a missed opportunity if the claim is central.
  • Figure ordering: Do the figures appear in the order of the paper's argument? Architecture → training → results → analysis is a natural arc. Figures out of narrative order confuse the reader's mental model.
  • Visual consistency across figures: Same method → same color/marker across all figures. Same data → same axis scale when comparison is relevant. Cross-reference with manuscript-review §12 (visual language consistency).
针对每个Figure，从8个维度进行评估：
维度1 — 论点-Figure一致性
核心问题：查看该Figure的读者是否能看到文本提出的论点？
  • 明确文本论点（例如：“方法A比方法B收敛更快”）
  • 明确视觉印象（例如：“两条几乎相同的曲线”）
  • 若两者不同：无论数据在技术上是否支持论点，该Figure都未能实现其修辞目的
失效模式：
  • 差异不可见：文本声称存在显著差异，但Figure的尺度、纵横比或数据密度使差异无法感知。数据存在，但视觉上无法体现。
  • 重点错误：Figure展示了很多内容，而文本仅讨论其中一项。读者不知道该关注哪里。
  • 印象矛盾：视觉印象主动暗示与论点相反的结论（例如：声称“收敛”但曲线仍在变化；声称“改进”但条形图看起来相等）。
  • 缺少上下文：Figure需要领域知识才能解读，但周围文本未提供该知识。
维度2 — 图表类型适配性
该可视化类型是否适合此论点？
论点类型有效图表无效图表
跨类别对比分组条形图、点图饼图、堆叠条形图（难以对比）
时间/序列趋势折线图条形图（掩盖连续性）
分布情况直方图、小提琴图、箱线图均值条形图（隐藏方差）
相关性/关系散点图配对值表格
构成/比例堆叠面积图、堆叠条形图多个饼图
差异/改进差异图、瀑布图全尺度并列条形图
排名水平条形图（排序后）垂直条形图（未排序）
空间分布热力图、等高线图数值表格
整体与部分树形图、堆叠条形图分组条形图
流程/流向桑基图、冲积图静态示意图
混淆/分类混淆矩阵（热力图）数值表格
消融实验贡献瀑布图表格或分组条形图
当存在更适合所提出论点的图表类型时，标记该问题。
维度3 — 坐标轴与尺度设计
坐标轴可以成就或破坏视觉论证。
  • Y轴起点：当差异较小时从0开始会使差异不可见。当展示绝对数量时从非0开始具有误导性。选择必须服务于论点：
    • 展示“我们的方法更优”→ 放大到相关范围
    • 展示“效果相对于总量较小”→ 从0开始
    • 始终：清晰标记坐标轴范围；必要时用
      //
      表示断裂轴
  • 坐标轴尺度：线性与对数。当数据跨越多个数量级或相对差异比绝对差异更重要时，对数尺度更合适。若线性尺度下数据跨越3个以上数量级，标记该问题。
  • 坐标轴范围：范围是否包含所有相关数据？是否超出数据范围过多（浪费空间）？选择范围是否为了夸大或最小化视觉差异？
  • 纵横比操纵：过宽或过窄的图表会夸大或平缓趋势。趋势线的斜率应可感知但不被夸大。
  • 双坐标轴：几乎总是令人困惑。两个不同的Y轴会引发错误的视觉对比。首选：两个独立面板，对齐X轴。
维度4 — 视觉层级与强调
Figure是否引导读者的视线到正确元素？
  • 主要元素：文本讨论的数据系列、区域或对比应在视觉上占主导（更粗的线条、更醒目的颜色、更大的标记、前景位置）。
  • 次要元素：上下文、基线和参考数据应在视觉上处于次要位置（更细的线条、灰色、更小的标记、背景）。
  • 注释：若文本引用特定点、区域或阈值，应在Figure中添加注释（箭头、标注、阴影区域、虚线参考线）。读者不必解码坐标来找到文本描述的内容。
  • 认知负荷：计算读者必须跟踪的不同视觉元素数量。单个图表中超过5-7个系列/组会超出工作记忆容量。拆分为面板或突出显示感兴趣的对比。
失效模式：
  • 所有线条/条形图具有相同的视觉权重——读者不知道什么是重要的
  • “我们的方法”线条与10条基线线条具有相同的宽度和样式
  • 文本称“注意第100轮的分歧”但Figure中没有任何标记第100轮的内容
  • 图例有15个条目——读者必须不断查阅
维度5 — 数据密度与简化
Figure展示的数据量是否适合其论点？
  • 过载：单个图表中有太多系列、类别或数据点。论点是关于两种方法之间的关系，但图表展示了12种方法。简化：突出显示感兴趣的对比；将其他方法移至补充材料或次要面板。
  • 不足：图表展示的数据太少，无法支撑论点。声称存在趋势但仅展示单个点。声称存在分布但仅展示两个条形图。声称存在收敛行为但仅展示三轮数据。
  • 需要原始数据时却展示汇总数据：仅展示均值但方差才是核心（或方差会削弱论点——两种情况都标记）。对于随机结果，应展示置信区间、误差线或小提琴图。
  • 需要汇总数据时却展示原始数据：声称存在聚合模式但展示单个数据点。展示10000个点的散点图，而密度图或分箱热力图更能展示结构。
维度6 — 图表说明作为解读指南
图表说明应告诉读者该看什么，而不仅仅是坐标轴是什么。
图表说明质量等级：
  1. 仅描述性（弱）：“五种方法的训练损失随轮次变化。”读者必须自行找出结论。
  2. 指导性（合格）：“训练损失随轮次变化。方法A（红色）约50轮收敛，而基线方法需要150+轮。”引导读者关注论点。
  3. 解释性（强）：“训练损失随轮次变化。方法A（红色）比最近的基线方法（蓝色）快3倍收敛，支持架构变更X降低优化难度的论点。”将视觉内容与论证联系起来。
针对每个图表说明，确定其等级，若为1级则建议升级。2级是有效沟通的最低要求。3级是支持核心论点的关键Figure的理想等级。
维度7 — 感知准确性
视觉编码是否准确表示基础数据？
  • 面积编码数量：若使用条形宽度、气泡大小或面积来编码数值，映射必须与面积成正比（而非半径或直径）。数值大2倍应对应面积大2倍，而非半径大2倍（对应面积大4倍）。
  • 3D效果：3D条形图、3D饼图、透视效果——这些会通过透视缩短扭曲数值感知。标记任何用3D可视化2D数据的情况。
  • 颜色尺度线性：顺序色图（viridis、plasma）具有感知上均匀的步长。彩虹/jet色图存在感知不连续性，会在连续数据中创建人为边界。标记用彩虹/jet色图展示连续数据的情况。
  • 未标记的截断轴：Y轴不从0开始且无可见断裂（
    //
    标记）会误导读者对相对大小的判断。
  • 纵横比失真：饼图不是圆形。条形宽度不一致。地图投影错误（在ML论文中少见，但在空间分析中常见）。
维度8 — 冗余与叙事逻辑
作为一个集合的Figures是否讲述连贯的故事？
  • 冗余Figures：两个Figures以不同形式展示基本相同的信息。除非每个都能提供独特见解，否则合并或选择更有效的一个。
  • 缺失Figures：文本中是否有核心论点没有视觉支持但会从中受益？如果论点是核心内容，缺少对应的Figure是错失机会。
  • Figure排序：Figures是否按论文论证的顺序出现？架构→训练→结果→分析是自然的逻辑顺序。Figures不符合叙事顺序会混淆读者的思维模型。
  • Figures间的视觉一致性：同一方法→在所有Figures中使用相同颜色/标记。同一数据→当需要对比时使用相同坐标轴尺度。参考manuscript-review §12（视觉语言一致性）。

3. Common Antipatterns

3. 常见反模式

Quick-reference of frequently occurring rhetorical failures:
AntipatternDescriptionFix
The Invisible WinMethod outperforms by 0.3% but y-axis spans 0-100%Zoom to relevant range; use difference plot
The Spaghetti Plot10+ overlapping lines, all same weightHighlight 2-3 of interest; gray out rest
The Bar Chart of MeansBars showing means without error bars/CIAdd error bars; consider violin/box plots
The Orphan ClaimText discusses a specific region/point with no annotationAdd arrow, shaded region, or reference line
The Pie ChartComparing proportions across >5 categoriesHorizontal bar chart, sorted
The Rainbow HeatmapJet/rainbow colormap for continuous dataUse perceptually uniform colormap (viridis)
The Giant LegendLegend with 15 entries reader must cross-referenceDirect labeling on lines; or reduce series count
The Wrong ChartLine chart for categorical data; bar chart for trendsMatch chart type to data type and claim
The Dual AxisTwo y-axes implying false correlationTwo separate panels, aligned x-axis
The Data DumpAll results in one figure "for completeness"Show what matters; appendix the rest
The Missing BaselineResults without visual reference pointAdd dashed line for baseline/random/human performance
The Abstract FigureText says "see Figure 3" but Figure 3 requires 5 minutes of studySimplify; annotate; upgrade caption
常见修辞失效速查：
反模式描述修复方案
隐形优势方法性能提升0.3%但Y轴范围为0-100%放大到相关范围；使用差异图
** spaghetti图**10+条重叠线条，权重相同突出显示2-3个感兴趣的线条；将其余线条设为灰色
均值条形图展示均值但无误差线/置信区间添加误差线；考虑使用小提琴图/箱线图
孤立论点文本讨论特定区域/点但无注释添加箭头、阴影区域或参考线
饼图对比超过5个类别的比例使用水平条形图并排序
彩虹热力图使用jet/彩虹色图展示连续数据使用感知均匀的色图（viridis）
巨型图例图例有15个条目，读者必须反复查阅在线条上直接标注；或减少系列数量
错误图表类型用折线图展示分类数据；用条形图展示趋势匹配图表类型与数据类型及论点
双坐标轴两个Y轴暗示错误相关性使用两个独立面板，对齐X轴
数据堆砌所有结果放在一个Figure中“以求完整”展示重要内容；其余放在附录
缺失基线结果无视觉参考点添加虚线表示基线/随机/人类性能
抽象Figure文本称“见图3”但图3需要5分钟才能理解简化；添加注释；升级图表说明

4. Generate Report

4. 生成报告

For each figure, produce:
markdown
undefined
针对每个Figure，生成：
markdown
undefined

Figure [N]: [brief description]

Figure [N]: [brief description]

Prose claim: [one-sentence claim the figure is supposed to support]
Visual takeaway: [what a reader actually sees at first glance]
Alignment: [STRONG | ADEQUATE | WEAK | CONTRADICTORY]
Issues:
  1. [Dimension]: [specific problem]
    • Impact: [how this affects the reader's understanding]
    • Fix: [specific recommendation with concrete changes]
Caption assessment: [Level 1/2/3] — [recommendation if upgrade needed]
Recommended changes: [prioritized list]

Then a summary section:

```markdown
Prose claim: [one-sentence claim the figure is supposed to support]
Visual takeaway: [what a reader actually sees at first glance]
Alignment: [STRONG | ADEQUATE | WEAK | CONTRADICTORY]
Issues:
  1. [Dimension]: [specific problem]
    • Impact: [how this affects the reader's understanding]
    • Fix: [specific recommendation with concrete changes]
Caption assessment: [Level 1/2/3] — [recommendation if upgrade needed]
Recommended changes: [prioritized list]

然后添加总结部分：

```markdown

Figure Set Assessment

Figure Set Assessment

Overall narrative coherence: [figures tell a coherent story / gaps exist / redundancies]
Strongest figure: Figure [N] — [why it works] Weakest figure: Figure [N] — [primary issue]
Priority fixes:
  1. [Most impactful change across all figures]
  2. [Second most impactful]
  3. [Third most impactful]
Missing figures: [claims that need visual support but lack it] Redundant figures: [figures that could be merged or cut]
undefined
Overall narrative coherence: [figures tell a coherent story / gaps exist / redundancies]
Strongest figure: Figure [N] — [why it works] Weakest figure: Figure [N] — [primary issue]
Priority fixes:
  1. [Most impactful change across all figures]
  2. [Second most impactful]
  3. [Third most impactful]
Missing figures: [claims that need visual support but lack it] Redundant figures: [figures that could be merged or cut]
undefined

5. Output

5. 输出

Save report as 
[manuscript-name]-figure-rhetoric-report.md
.
Present:
  • Count of figures by alignment rating (STRONG / ADEQUATE / WEAK / CONTRADICTORY)
  • Top 3 fixes by impact
  • Any CONTRADICTORY figures (highest priority — these actively hurt the paper)
将报告保存为
[manuscript-name]-figure-rhetoric-report.md
。
展示：
  • 按一致性评级（STRONG / ADEQUATE / WEAK / CONTRADICTORY）划分的Figure数量
  • 影响最大的3项修复
  • 任何CONTRADICTORY级别的Figures（最高优先级——这些会严重损害论文）

Core Principles

核心原则

  • The reader is naive. Do not evaluate figures through the author's eyes. The author knows what the figure "should" show. The reader sees only what is visually present. Every judgment in this audit is from the reader's perspective.
  • Claim first, then figure. Read the prose claim before looking at the figure. The figure's job is to support that specific claim. If the figure is beautiful but doesn't support the claim, it fails.
  • One figure, one message. A figure trying to show three things shows none of them clearly. If the text makes three claims about one figure, the figure is overloaded or the text should point to three figures.
  • Visual perception trumps data accuracy. A figure can be numerically correct but perceptually wrong (e.g., differences invisible due to scale). The reader's visual impression IS the communication. If the impression doesn't match the claim, the figure has failed.
  • Concreteness over abstraction. Recommendations specify the exact change: "change y-axis range from 0-100 to 85-95," not "consider adjusting the axis." Include suggested chart types, axis ranges, color choices, and annotation text.
  • Severity scales with claim importance. A weak figure supporting a minor methodological point is LOW priority. A weak figure supporting the paper's core result is CRITICAL — it's the first thing a reviewer scrutinizes.
  • 读者是普通读者。不要从作者的视角评估Figures。作者知道Figure“应该”展示什么。读者只能看到视觉上呈现的内容。本次审计的所有判断均从读者视角出发。
  • 先论点，后Figure。在查看Figure之前先阅读文本论点。Figure的职责是支持该特定论点。如果Figure很漂亮但不支持论点，它就是失败的。
  • 一个Figure，一个信息。试图展示三件事的Figure无法清晰展示任何一件。如果文本对一个Figure提出三个论点，要么Figure过载，要么文本应指向三个Figures。
  • 视觉感知优于数据准确性。一个Figure在数值上正确但感知上错误（例如：因尺度导致差异不可见）。读者的视觉印象就是传达的内容。如果印象与论点不匹配，Figure就失败了。
  • 具体而非抽象。建议明确具体的更改：“将Y轴范围从0-100调整为85-95”，而非“考虑调整坐标轴”。包括建议的图表类型、坐标轴范围、颜色选择和注释文本。
  • 严重程度随论点重要性递增。支持次要方法论观点的弱Figure优先级低。支持论文核心结果的弱Figure优先级极高——这是审稿人首先会仔细审查的内容。