paper-write

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Paper Write: Section-by-Section LaTeX Generation

论文撰写：逐节LaTeX生成

Draft a LaTeX paper based on: $ARGUMENTS

根据**$ARGUMENTS**撰写LaTeX论文

Constants

常量

REVIEWER_MODEL =
gpt-5.4
— Model used via Codex MCP for section review. Must be an OpenAI model.
TARGET_VENUE =
ICLR
— Default venue. Supported:
```
ICLR
```
,
```
NeurIPS
```
,
```
ICML
```
. Determines style file and formatting.
ANONYMOUS = true — If true, use anonymous author block. Set
```
false
```
for camera-ready.
MAX_PAGES = 9 — Main body page limit. Counts from first page to end of Conclusion section. References and appendix are NOT counted.
DBLP_BIBTEX = true — Fetch real BibTeX from DBLP/CrossRef instead of LLM-generated entries. Eliminates hallucinated citations. Zero install required. Set
```
false
```
to use legacy behavior (LLM search +
```
[VERIFY]
```
markers).

REVIEWER_MODEL =
gpt-5.4
— 通过Codex MCP用于章节评审的模型，必须是OpenAI模型。
TARGET_VENUE =
ICLR
— 默认会议场地。支持：
```
ICLR
```
、
```
NeurIPS
```
、
```
ICML
```
，决定样式文件和格式。
ANONYMOUS = true — 如果为true，使用匿名作者块。终稿时设置为
```
false
```
。
MAX_PAGES = 9 — 主体内容页数限制，从第一页到结论章节结束。参考文献和附录不计入。
DBLP_BIBTEX = true — 从DBLP/CrossRef获取真实的BibTeX，而非大语言模型生成的条目，消除虚构引用。无需安装，设置为
```
false
```
则使用旧版逻辑（大语言模型搜索 +
```
[VERIFY]
```
标记）。

Inputs

输入

PAPER_PLAN.md — outline with claims-evidence matrix, section plan, figure plan (from
```
/paper-plan
```
)
NARRATIVE_REPORT.md — the research narrative (primary source of content)
Generated figures — PDF/PNG files in
```
figures/
```
(from
```
/paper-figure
```
)
LaTeX includes —
```
figures/latex_includes.tex
```
(from
```
/paper-figure
```
)
Bibliography — existing
```
.bib
```
file, or will create one

If no PAPER_PLAN.md exists, ask the user to run

/paper-plan

first or provide a brief outline.

PAPER_PLAN.md — 包含论点-证据矩阵、章节规划、图表规划的大纲（来自
```
/paper-plan
```
）
NARRATIVE_REPORT.md — 研究叙述（内容的主要来源）
生成的图表 —
```
figures/
```
目录下的PDF/PNG文件（来自
```
/paper-figure
```
）
LaTeX引用文件 —
```
figures/latex_includes.tex
```
（来自
```
/paper-figure
```
）
参考文献 — 已有的
```
.bib
```
文件，或将自动创建新文件

如果不存在PAPER_PLAN.md，请先让用户运行

/paper-plan

或提供简要大纲。

Templates

模板

Venue-Specific Setup

特定会议场地配置

The skill includes conference templates in

templates/

. Select based on TARGET_VENUE:

ICLR:

latex

\documentclass{article}
\usepackage{iclr2026_conference,times}
% \iclrfinalcopy  % Uncomment for camera-ready

NeurIPS:

latex

\documentclass{article}
\usepackage[preprint]{neurips_2025}
% \usepackage[final]{neurips_2025}  % Camera-ready

ICML:

latex

\documentclass[accepted]{icml2025}
% Use [accepted] for camera-ready

本技能在

templates/

目录中包含会议模板，根据TARGET_VENUE选择：

ICLR:

latex

\documentclass{article}
\usepackage{iclr2026_conference,times}
% \iclrfinalcopy  % 终稿时取消注释

NeurIPS:

latex

\documentclass{article}
\usepackage[preprint]{neurips_2025}
% \usepackage[final]{neurips_2025}  % 终稿版本

ICML:

latex

\documentclass[accepted]{icml2025}
% 终稿使用[accepted]参数

Project Structure

项目结构

Generate this file structure:

paper/
├── main.tex                    # master file (includes sections)
├── iclr2026_conference.sty     # or neurips_2025.sty / icml2025.sty
├── math_commands.tex           # shared math macros
├── references.bib              # bibliography (filtered — only cited entries)
├── sections/
│   ├── 0_abstract.tex
│   ├── 1_introduction.tex
│   ├── 2_related_work.tex
│   ├── 3_method.tex            # or preliminaries, setup, etc.
│   ├── 4_experiments.tex
│   ├── 5_conclusion.tex
│   └── A_appendix.tex          # proof details, extra experiments
└── figures/                    # symlink or copy from project figures/

Section files are FLEXIBLE: If the paper plan has 6-8 sections, create corresponding files (e.g.,

4_theory.tex

5_experiments.tex

6_analysis.tex

7_conclusion.tex

生成以下文件结构：

paper/
├── main.tex                    # 主文件（引用各章节）
├── iclr2026_conference.sty     # 或neurips_2025.sty / icml2025.sty
├── math_commands.tex           # 共享数学宏定义
├── references.bib              # 参考文献（仅包含被引用的条目）
├── sections/
│   ├── 0_abstract.tex
│   ├── 1_introduction.tex
│   ├── 2_related_work.tex
│   ├── 3_method.tex            # 或 preliminaries、setup等
│   ├── 4_experiments.tex
│   ├── 5_conclusion.tex
│   └── A_appendix.tex          # 证明细节、额外实验
└── figures/                    # 符号链接或复制自项目的figures/

章节文件可灵活调整：如果论文规划有6-8个章节，创建对应的文件（例如

4_theory.tex

、

5_experiments.tex

、

6_analysis.tex

、

7_conclusion.tex

）。

Workflow

工作流

Step 0: Backup and Clean

步骤0：备份与清理

paper/

already exists, back up to

paper-backup-{timestamp}/

before overwriting. Never silently destroy existing work.

CRITICAL: Clean stale files. When changing section structure (e.g., 5 sections → 7 sections), delete section files that are no longer referenced by

main.tex

. Stale files (e.g., old

5_conclusion.tex

left behind when conclusion moved to

7_conclusion.tex

) cause confusion and waste space.

如果

paper/

目录已存在，在覆盖前先备份到

paper-backup-{timestamp}/

目录。绝不静默删除现有工作成果。

重要提示：清理过期文件。当章节结构变更时（例如从5章变为7章），删除

main.tex

不再引用的章节文件。过期文件（例如结论章节移到

7_conclusion.tex

后遗留的旧

5_conclusion.tex

）会造成混淆并占用空间。

Step 1: Initialize Project

步骤1：初始化项目

Create
```
paper/
```
directory
Copy venue template from
```
templates/
```
— the template already includes:
- All standard packages (amsmath, hyperref, cleveref, booktabs, etc.)
- Theorem environments with
```
\crefname{assumption}
```
  fix
- Anonymous author block
Generate
```
math_commands.tex
```
with paper-specific notation
Create section files matching PAPER_PLAN structure

Author block (anonymous mode):

latex

\author{Anonymous Authors}

创建
```
paper/
```
目录
从
```
templates/
```
复制会议场地模板 — 模板已包含：
- 所有标准包（amsmath、hyperref、cleveref、booktabs等）
- 修复了
```
\crefname{assumption}
```
  的定理环境
- 匿名作者块
根据论文专属符号生成
```
math_commands.tex
```
创建与PAPER_PLAN结构匹配的章节文件

匿名模式作者块：

latex

\author{Anonymous Authors}

Step 2: Generate math_commands.tex

步骤2：生成math_commands.tex

Create shared math macros based on the paper's notation:

latex

% math_commands.tex — shared notation
\newcommand{\R}{\mathbb{R}}
\newcommand{\E}{\mathbb{E}}
\DeclareMathOperator*{\argmin}{arg\,min}
\DeclareMathOperator*{\argmax}{arg\,max}
% Add paper-specific notation here

根据论文符号创建共享数学宏定义：

latex

% math_commands.tex — 共享符号定义
\newcommand{\R}{\mathbb{R}}
\newcommand{\E}{\mathbb{E}}
\DeclareMathOperator*{\argmin}{arg\,min}
\DeclareMathOperator*{\argmax}{arg\,max}
% 在此添加论文专属符号

Step 3: Write Each Section

步骤3：撰写各章节

Process sections in order. For each section:

Read the plan — what claims, evidence, citations belong here
Read NARRATIVE_REPORT.md — extract relevant content, findings, and quantitative results
Draft content — write complete LaTeX (not placeholders)
Insert figures/tables — use snippets from
```
figures/latex_includes.tex
```
Add citations — use
```
\citep{}
```
/
```
\citet{}
```
(all three venues use
```
natbib
```
)

按顺序处理每个章节。针对每个章节：

阅读规划 — 明确本章节应包含的论点、证据和引用
阅读NARRATIVE_REPORT.md — 提取相关内容、研究发现和量化结果
撰写内容 — 编写完整的LaTeX代码（而非占位符）
插入图表/表格 — 使用
```
figures/latex_includes.tex
```
中的代码片段
添加引用 — 使用
```
\citep{}
```
/
```
\citet{}
```
（三个会议均使用
```
natbib
```
）

Section-Specific Guidelines

章节专属指南

§0 Abstract:

Must be self-contained (understandable without reading the paper)
Structure: problem → approach → key result → implication
Include one concrete quantitative result
150-250 words (check venue limit)
No citations, no undefined acronyms
No
```
\begin{abstract}
```
— that's in main.tex

§1 Introduction:

Open with a compelling hook (1-2 sentences, problem motivation)
State the gap clearly ("However, ...")
List contributions as a numbered or bulleted list
End with a brief roadmap ("The rest of this paper is organized as...")
Include the main result figure if space allows
Target: 1.5 pages

§2 Related Work:

MINIMUM 1 full page (3-4 substantive paragraphs). Short related work sections are a common reviewer complaint.
Organize by category using
```
\paragraph{Category Name.}
```
Each category: 1 paragraph summarizing the line of work + 1-2 sentences positioning this paper
Do NOT just list papers — synthesize and compare
End each paragraph with how this paper relates/differs

§3 Method / Preliminaries / Setup:

Define notation early (reference math_commands.tex)
Use
```
\begin{definition}
```
,
```
\begin{theorem}
```
environments for formal statements
For theory papers: include proof sketches of key results in main body, full proofs in appendix
For theory papers: include a comparison table of prior bounds vs. this paper
Include algorithm pseudocode if applicable (
```
algorithm2e
```
or
```
algorithmic
```
)
Target: 1.5-2 pages

§4 Experiments:

Start with experimental setup (datasets, baselines, metrics, implementation details)
Main results table/figure first
Then ablations and analysis
Every claim from the introduction must have supporting evidence here
Target: 2.5-3 pages

§5 Conclusion:

Summarize contributions (NOT copy-paste from intro — rephrase)
Limitations (be honest — reviewers appreciate this)
Future work (1-2 concrete directions)
Ethics statement and reproducibility statement (if venue requires)
Target: 0.5 pages

Appendix:

Proof details (full proofs of main-body theorems)
Additional experiments, ablations
Implementation details, hyperparameter tables
Additional visualizations

§0 摘要：

必须独立成篇（无需阅读论文即可理解）
结构：问题→方法→核心结果→意义
包含一项具体的量化结果
150-250词（需符合场地限制）
无引用、无未定义的缩写
无需
```
\begin{abstract}
```
标签 — 该标签已在main.tex中

§1 引言：

以引人入胜的切入点开篇（1-2句话，阐述问题动机）
清晰指出研究空白（“然而，……”）
以编号或项目符号列表列出研究贡献
结尾简要介绍论文结构（“本文剩余部分安排如下……”）
若空间允许，加入核心结果图表
目标篇幅：1.5页

§2 相关工作：

至少1整页（3-4个实质性段落）。简短的相关工作章节是审稿人常见的槽点。
使用
```
\paragraph{分类名称。}
```
按类别组织内容
每个类别：1段总结该研究方向 + 1-2句话定位本文
不要仅罗列论文 — 要进行综合与对比
每个段落结尾说明本文与该方向的关联/差异

§3 方法/预备知识/实验设置：

尽早定义符号（引用math_commands.tex）
使用
```
\begin{definition}
```
、
```
\begin{theorem}
```
环境撰写正式声明
理论类论文：在主体部分包含核心结果的证明梗概，完整证明放在附录
理论类论文：包含现有边界与本文结果的对比表格
若适用，添加算法伪代码（使用
```
algorithm2e
```
或
```
algorithmic
```
）
目标篇幅：1.5-2页

§4 实验：

从实验设置开始（数据集、基线模型、指标、实现细节）
先展示核心结果表格/图表
再展示消融实验与分析
引言中的每个论点都必须在此有支撑证据
目标篇幅：2.5-3页

§5 结论：

总结研究贡献（不要直接复制引言内容 — 需重新表述）
研究局限（如实说明 — 审稿人会认可这一点）
未来工作方向（1-2个具体方向）
伦理声明与可复现性声明（若场地要求）
目标篇幅：0.5页

附录：

证明细节（主体部分定理的完整证明）
额外实验、消融实验
实现细节、超参数表格
额外可视化内容

Step 4: Build Bibliography

步骤4：构建参考文献

CRITICAL: Only include entries that are actually cited in the paper.

Scan all
```
\citep{}
```
and
```
\citet{}
```
references in the drafted sections
Build a citation key list
For each citation key:
- Check existing
```
.bib
```
  files in the project/narrative docs
- If not found and DBLP_BIBTEX = true, use the verified fetch chain below
- If not found and DBLP_BIBTEX = false, search arXiv/Scholar for correct BibTeX
- NEVER fabricate BibTeX entries — mark unknown ones with
```
[VERIFY]
```
  comment
Write
```
references.bib
```
containing ONLY cited entries (no bloat)

重要提示：仅包含论文中实际被引用的条目

扫描所有已撰写章节中的
```
\citep{}
```
和
```
\citet{}
```
引用
生成引用键列表
针对每个引用键：
- 检查项目/研究叙述文档中已有的
```
.bib
```
  文件
- 若未找到且DBLP_BIBTEX = true，使用以下验证获取链
- 若未找到且DBLP_BIBTEX = false，在arXiv/学术搜索中查找正确的BibTeX
- 绝不虚构BibTeX条目 — 未知条目需标记
```
[VERIFY]
```
  注释
编写仅包含被引用条目的
```
references.bib
```
（无冗余内容）

Verified BibTeX Fetch (when DBLP_BIBTEX = true)

验证式BibTeX获取（当DBLP_BIBTEX = true时）

Three-step fallback chain — zero install, zero auth, all real BibTeX:

Step A: DBLP (best quality — full venue, pages, editors)

bash

undefined

三步 fallback 链 — 无需安装、无需认证，所有BibTeX均为真实内容：

步骤A：DBLP（质量最佳 — 包含完整场地、页码、编辑信息）

bash

undefined

1. Search by title + first author

1. 按标题+第一作者搜索

curl -s "https://dblp.org/search/publ/api?q=TITLE+AUTHOR&format=json&h=3"

2. Extract DBLP key from result (e.g., conf/nips/VaswaniSPUJGKP17)

2. 从结果中提取DBLP键（例如 conf/nips/VaswaniSPUJGKP17）

3. Fetch real BibTeX

3. 获取真实BibTeX

curl -s "https://dblp.org/rec/{key}.bib"


**Step B: CrossRef DOI (fallback — works for arXiv preprints)**
```bash

curl -s "https://dblp.org/rec/{key}.bib"


**步骤B：CrossRef DOI（备选方案 — 适用于arXiv预印本）**
```bash

If paper has a DOI or arXiv ID (arXiv DOI = 10.48550/arXiv.{id})

如果论文有DOI或arXiv ID（arXiv DOI格式为 10.48550/arXiv.{id}）

curl -sLH "Accept: application/x-bibtex" "https://doi.org/{doi}"


**Step C: Mark `[VERIFY]` (last resort)**
If both DBLP and CrossRef return nothing, mark the entry with `% [VERIFY]` comment. Do NOT fabricate.

**Why this matters:** LLM-generated BibTeX frequently hallucinates venue names, page numbers, or even co-authors. DBLP and CrossRef return publisher-verified metadata. Upstream skills (`/research-lit`, `/novelty-check`) may mention papers from LLM memory — this fetch chain is the gate that prevents hallucinated citations from entering the final `.bib`.

**Automated bib cleaning** — use this Python pattern to extract only cited entries:

```python
import re

curl -sLH "Accept: application/x-bibtex" "https://doi.org/{doi}"


**步骤C：标记`[VERIFY]`（最后手段）**
如果DBLP和CrossRef均无结果，在条目上标记`% [VERIFY]`注释。绝不虚构内容。

**为何如此重要：** 大语言模型生成的BibTeX经常虚构场地名称、页码甚至合著者。DBLP和CrossRef返回的是出版商验证的元数据。上游技能（`/research-lit`、`/novelty-check`）可能会提及大语言模型记忆中的论文 — 该获取链是防止虚构引用进入最终`.bib`文件的关卡。

**自动化参考文献清理** — 使用以下Python模式提取仅被引用的条目：

```python
import re

1. Grep all \citep{...} and \citet{...} from all .tex files

1. 从所有.tex文件中提取所有\citep{...}和\citet{...}

2. Extract unique keys (handle multi-cite like \citep{a,b,c})

2. 提取唯一引用键（处理\citep{a,b,c}这类多引用情况）

3. Parse the full .bib file, keep only entries whose key is in the cited set

3. 解析完整.bib文件，仅保留引用键在被引用集合中的条目

4. Write the filtered bib

4. 写入过滤后的参考文献文件


This prevents bib bloat (e.g., 948 lines → 215 lines in testing).

**Citation verification rules (from claude-scholar + Imbad0202):**
1. Every BibTeX entry must have: author, title, year, venue/journal
2. Prefer published venue versions over arXiv preprints (if published)
3. Use consistent key format: `{firstauthor}{year}{keyword}` (e.g., `ho2020denoising`)
4. Double-check year and venue for every entry
5. Remove duplicate entries (same paper with different keys)


这能避免参考文献冗余（测试中可从948行缩减至215行）。

Step 5: De-AI Polish (from kgraph57/paper-writer-skill)

步骤5：去AI化润色（来自kgraph57/paper-writer-skill）

After drafting all sections, scan for common AI writing patterns and fix them:

Content patterns to fix:

Significance inflation ("groundbreaking", "revolutionary" → use measured language)
Formulaic transitions ("In this section, we..." → remove or vary)
Generic conclusions ("This work opens exciting new avenues" → be specific)

Language patterns to fix (watch words):

Replace: delve, pivotal, landscape, tapestry, underscore, noteworthy, intriguingly
Remove filler: "It is worth noting that", "Importantly,", "Notably,"
Avoid rule-of-three lists ("X, Y, and Z" appearing repeatedly)
Don't start consecutive sentences with "This" or "We"

完成所有章节撰写后，扫描常见AI写作模式并修正：

需修正的内容模式：

夸大重要性（将“突破性的”“革命性的”改为平实表述）
公式化过渡（删除或替换“在本节中，我们……”这类表述）
通用结论（将“本工作开辟了令人兴奋的新方向”改为具体表述）

需修正的语言模式（警惕词汇）：

替换：delve、pivotal、landscape、tapestry、underscore、noteworthy、intriguingly
删除冗余表述：“值得注意的是”“重要的是”“值得一提的是”
避免三连列表（重复出现“X、Y和Z”结构）
不要连续用“这”或“我们”开头

Step 6: Cross-Review with REVIEWER_MODEL

步骤6：使用REVIEWER_MODEL交叉评审

Send the complete draft to GPT-5.4 xhigh:

mcp__codex__codex:
  model: gpt-5.4
  config: {"model_reasoning_effort": "xhigh"}
  prompt: |
    Review this [VENUE] paper draft (main body, excluding appendix).

    Focus on:
    1. Does each claim from the intro have supporting evidence?
    2. Is the writing clear, concise, and free of AI-isms?
    3. Any logical gaps or unclear explanations?
    4. Does it fit within [MAX_PAGES] pages (to end of Conclusion)?
    5. Is related work sufficiently comprehensive (≥1 page)?
    6. For theory papers: are proof sketches adequate?
    7. Are figures/tables clearly described and properly referenced?

    For each issue, specify: severity (CRITICAL/MAJOR/MINOR), location, and fix.

    [paste full draft text]

Apply CRITICAL and MAJOR fixes. Document MINOR issues for the user.

将完整草稿发送给GPT-5.4 xhigh：

mcp__codex__codex:
  model: gpt-5.4
  config: {"model_reasoning_effort": "xhigh"}
  prompt: |
    评审这份[VENUE]论文草稿（主体内容，不含附录）。

    重点关注：
    1. 引言中的每个论点是否有支撑证据？
    2. 写作是否清晰、简洁且无AI腔？
    3. 是否存在逻辑空白或解释不清的地方？
    4. 篇幅是否控制在[MAX_PAGES]页内（到结论章节结束）？
    5. 相关工作是否足够全面（≥1页）？
    6. 理论类论文：证明梗概是否充分？
    7. 图表是否描述清晰且引用正确？

    针对每个问题，明确：严重程度（CRITICAL/MAJOR/MINOR）、位置和修正方案。

    [粘贴完整草稿文本]

应用CRITICAL和MAJOR级别的修正。将MINOR级别的问题记录下来告知用户。

Step 7: Reverse Outline Test (from Research-Paper-Writing-Skills)

步骤7：反向大纲测试（来自Research-Paper-Writing-Skills）

After drafting all sections:

Extract topic sentences — pull the first sentence of every paragraph
Read them in sequence — they should form a coherent narrative on their own
Check claim coverage — every claim from the Claims-Evidence Matrix must appear
Check evidence mapping — every experiment/figure must support a stated claim
Fix gaps — if a topic sentence doesn't advance the story, rewrite the paragraph

完成所有章节撰写后：

提取主题句 — 提取每个段落的第一句话
按顺序阅读 — 这些句子本身应能构成连贯的叙述
检查论点覆盖 — 论点-证据矩阵中的每个论点都必须出现
检查证据映射 — 每个实验/图表都必须支撑一个明确的论点
填补空白 — 如果某个主题句无法推进叙述，重写该段落

Step 8: Final Checks

步骤8：最终检查

Before declaring done:

All
```
\ref{}
```
and
```
\label{}
```
match (no undefined references)
All
```
\citep{}
```
/
```
\citet{}
```
have corresponding BibTeX entries
No author information in anonymous mode
Figure/table numbering is correct
Page count within MAX_PAGES (main body to Conclusion end)
No TODO/FIXME/XXX markers left in the text
No
```
[VERIFY]
```
markers left unchecked
Abstract is self-contained (understandable without reading the paper)
Title is specific and informative (not generic)
Related work is ≥1 full page
references.bib contains ONLY cited entries (no bloat)
No stale section files — every .tex in
```
sections/
```
is
```
\input
```
ed by
```
main.tex
```
Section files match main.tex — file numbering and
```
\input
```
paths are consistent

在完成前确认：

所有
```
\ref{}
```
和
```
\label{}
```
匹配（无未定义引用）
所有
```
\citep{}
```
/
```
\citet{}
```
都有对应的BibTeX条目
匿名模式下无作者信息
图表编号正确
主体篇幅在MAX_PAGES限制内（到结论章节结束）
文本中无TODO/FIXME/XXX标记
无未检查的
```
[VERIFY]
```
标记
摘要独立成篇（无需阅读论文即可理解）
标题具体且信息丰富（非通用表述）
相关工作篇幅≥1整页
references.bib仅包含被引用的条目（无冗余）
无过期章节文件 —
```
sections/
```
中的每个.tex文件都被
```
main.tex
```
通过
```
\input
```
引用
章节文件与main.tex一致 — 文件编号和
```
\input
```
路径保持一致

Key Rules

核心规则

Large file handling: If the Write tool fails due to file size, immediately retry using Bash (
```
cat << 'EOF' > file
```
) to write in chunks. Do NOT ask the user for permission — just do it silently.
Do NOT generate author names, emails, or affiliations — use anonymous block or placeholder
Write complete sections, not outlines — the output should be compilable LaTeX
One file per section — modular structure for easy editing
Every claim must cite evidence — cross-reference the Claims-Evidence Matrix
Compile-ready — the output should compile with
```
latexmk
```
without errors (modulo missing figures)
No over-claiming — use hedging language ("suggests", "indicates") for weak evidence
Venue style matters — all three venues (ICLR/NeurIPS/ICML) use
```
natbib
```
(
```
\citep
```
/
```
\citet
```
)
Page limit = main body to Conclusion — references and appendix do NOT count
Clean bib — references.bib must only contain entries that are actually
```
\cite
```
d
Section count is flexible — match PAPER_PLAN structure, don't force into 5 sections
Backup before overwrite — never destroy existing
```
paper/
```
directory without backing up

大文件处理：若写入工具因文件大小失败，立即使用Bash（
```
cat << 'EOF' > file
```
）分块写入重试。无需询问用户许可 — 直接静默执行。
绝不生成作者姓名、邮箱或所属机构 — 使用匿名块或占位符
撰写完整章节，而非大纲 — 输出应为可编译的LaTeX代码
一章一文件 — 模块化结构便于编辑
每个论点必须引用证据 — 交叉引用论点-证据矩阵
可直接编译 — 输出应能通过
```
latexmk
```
编译（除缺失图表外无错误）
绝不夸大成果 — 对弱证据使用谨慎表述（“表明”“显示”）
场地样式至关重要 — 三个会议（ICLR/NeurIPS/ICML）均使用
```
natbib
```
（
```
\citep
```
/
```
\citet
```
）
页数限制 = 主体到结论章节 — 参考文献和附录不计入
精简参考文献 — references.bib必须仅包含实际被
```
\cite
```
引用的条目
章节数量灵活 — 匹配PAPER_PLAN结构，不要强行分为5章
覆盖前先备份 — 绝不未备份就删除现有
```
paper/
```
目录

Writing Quality Reference

写作质量参考

Principles from Research-Paper-Writing-Skills:

One message per paragraph — each paragraph makes exactly one point
Topic sentence first — the first sentence states the paragraph's message
Explicit transitions — connect paragraphs with logical connectors
Reverse outline test — extract topic sentences; they should form a coherent narrative

De-AI patterns from kgraph57/paper-writer-skill:

No AI watch words — delve, pivotal, landscape, tapestry, underscore
No significance inflation — groundbreaking, revolutionary, paradigm shift
No formulaic structures — vary sentence openings and transitions

来自Research-Paper-Writing-Skills的原则：

每段仅传递一个信息 — 每个段落仅表达一个核心观点
主题句前置 — 第一句话点明段落核心
显式过渡 — 使用逻辑连接词衔接段落
反向大纲测试 — 提取主题句；它们应能构成连贯的叙述

来自kgraph57/paper-writer-skill的去AI化模式：

避免AI特征词汇 — delve、pivotal、landscape、tapestry、underscore
不夸大重要性 — 避免使用“突破性的”“革命性的”“范式转变”等表述
避免公式化结构 — 变换句子开头和过渡方式

Acknowledgements

致谢

Writing methodology adapted from Research-Paper-Writing-Skills (CCF award-winning methodology). Citation verification from claude-scholar and Imbad0202/academic-research-skills. De-AI polish from kgraph57/paper-writer-skill. Backup mechanism from baoyu-skills.

写作方法改编自Research-Paper-Writing-Skills（获CCF奖项的方法）。引用验证来自claude-scholar和Imbad0202/academic-research-skills。去AI化润色来自kgraph57/paper-writer-skill。备份机制来自baoyu-skills。