paper-write

name	paper-write
description	Draft paper content section by section from a paper plan, first as complete Markdown drafts and then as one LaTeX file per section. Use when user says \"写论文\", \"write paper\", \"draft LaTeX\", \"开始写\", wants Markdown-first paper drafting, or wants to generate LaTeX content from a problem-aligned paper plan.

Paper Write: Markdown-First Section Generation

Draft a paper based on: $ARGUMENTS. Write complete Markdown drafts first, then convert the approved content into compile-ready LaTeX section files.

Constants

• REVIEW_ROLE = paper architect/reviewer — Use the AGENTS.md paper review role for draft diagnosis. Do not hard-code a model.
• TARGET_VENUE = ICLR — Default venue. Supported: ICLR, NeurIPS, ICML, CVPR (also ICCV/ECCV), ACL (also EMNLP/NAACL), AAAI, ACM (ACM MM, SIGIR, KDD, CHI, etc.), IEEE_JOURNAL (IEEE Transactions / Letters, e.g., T-PAMI, JSAC, TWC, TCOM, TSP, TIP), IEEE_CONF (IEEE conferences, e.g., ICC, GLOBECOM, INFOCOM, ICASSP). Determines style file and formatting.
• ANONYMOUS = true — If true, use anonymous author block. Set false for camera-ready. Note: most IEEE venues do NOT use anonymous submission — set false for IEEE.
• MAX_PAGES = 9 — Main body page limit. For ML conferences: counts from first page to end of Conclusion section, references and appendix NOT counted. For IEEE venues: references ARE counted toward the page limit. Typical limits: IEEE journal = no strict limit (but 12-14 pages typical for Transactions, 4-5 for Letters), IEEE conference = 5-8 pages including references.
• 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).

Inputs

1. PAPER_PLAN.md — outline with claims-evidence matrix, section plan, figure plan (from /paper-plan)
   • Prefer plans that include Problem Definition Lock, Problem-Story Alignment Matrix, and Introduction Blueprint
2. NARRATIVE_REPORT.md — the research narrative (primary source of content)
3. Generated figures — PDF/PNG files in figures/ (from /paper-figure)
4. LaTeX includes — figures/latex_includes.tex (from /paper-figure)
5. 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.

Orchestra-Guided Writing Overlay

Keep the existing workflow, file layout, and defaults. Use the shared references below only when they improve writing quality:

• Read ../shared-references/writing-principles.md before drafting the Abstract, Introduction, Related Work, or when prose feels generic
• Read ../shared-references/venue-checklists.md during the final write-up and submission-readiness pass
• Read ../shared-references/citation-discipline.md only when the built-in DBLP/CrossRef workflow is insufficient

These references are support material, not extra workflow phases.

Templates

Venue-Specific Setup

The skill includes conference templates in templates/. Select based on TARGET_VENUE:

ICLR:

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

NeurIPS:

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

ICML:

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

IEEE Journal (Transactions, Letters):

\documentclass[journal]{IEEEtran}
\usepackage{cite}  % IEEE uses \cite{}, NOT natbib
% Author block uses \author{Name~\IEEEmembership{Member,~IEEE}}

IEEE Conference (ICC, GLOBECOM, INFOCOM, ICASSP, etc.):

\documentclass[conference]{IEEEtran}
\usepackage{cite}  % IEEE uses \cite{}, NOT natbib
% Author block uses \IEEEauthorblockN / \IEEEauthorblockA

Project Structure

Generate this file structure:

paper/
├── main.tex                    # master file (includes sections)
├── iclr2026_conference.sty     # or neurips_2025.sty / icml2025.sty / IEEEtran.cls + IEEEtran.bst
├── 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
├── drafts/
│   ├── 0_abstract.md
│   ├── 1_introduction.md
│   ├── 2_related_work.md
│   ├── 3_method.md
│   ├── 4_experiments.md
│   └── 5_conclusion.md
└── figures/                    # symlink or copy from project figures/

Section files are FLEXIBLE: If the paper plan has 6-8 sections, create corresponding Markdown drafts and LaTeX files (e.g., 4_theory.md/.tex, 5_experiments.md/.tex, 6_analysis.md/.tex, 7_conclusion.md/.tex).

Workflow

Step 0: Backup and Clean

If paper/ already exists, back up to paper-backup-{timestamp}/ before overwriting. Never silently destroy existing work.

CRITICAL: Handle stale files safely. When changing section structure (e.g., 5 sections → 7 sections), list section files that are no longer referenced by main.tex. Archive or delete only files proven to be generated by the current skill run; otherwise ask before removing them.

Step 1: Initialize Project

1. Create paper/ directory
2. 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
3. Generate math_commands.tex with paper-specific notation
4. Create paper/drafts/ and paper/sections/ files matching PAPER_PLAN structure

Author block (anonymous mode):

\author{Anonymous Authors}

Step 2: Generate math_commands.tex

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

% 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

Step 3: Write Markdown Drafts First

Process sections in order. For each section, first write a complete Markdown draft in paper/drafts/. These drafts are not outlines; they are the full paper content in plain prose with figure/table placeholders and citation keys.

1. Read the plan — what claims, evidence, citations belong here
2. Read NARRATIVE_REPORT.md — extract relevant content, findings, and quantitative results
3. Use problem alignment — map each section to Problem Definition Lock, Problem-Story Alignment Matrix, and the Claims-Evidence Matrix
4. Draft Markdown — write complete prose, not bullet placeholders
5. Add evidence placeholders — mark planned figure/table locations and citation keys

Each Markdown draft must include a short hidden planning block at the top:

<!--
Problem Alignment Notes:
- Problem IDs covered:
- Contributions supported:
- Evidence used:
- Claims not yet supported:
-->

If Problem Definition Lock or Introduction Blueprint is missing from PAPER_PLAN.md, infer a draft version from the user's materials and pause only if multiple incompatible problem framings would change the paper's claims.

Step 4: Convert Markdown Drafts to LaTeX Section Files

After the Markdown drafts are complete and internally consistent, convert each draft to one LaTeX file in paper/sections/.

1. Preserve the Markdown draft's technical content and paragraph order unless LaTeX formatting requires minor restructuring.
2. Replace Markdown figure/table placeholders with snippets from figures/latex_includes.tex.
3. Add citations — for ML conferences (ICLR/NeurIPS/ICML/CVPR/ACL/AAAI): use \citep{} / \citet{} (natbib). For IEEE venues: use \cite{} (numeric style via cite package). Never mix natbib and cite commands.
4. Keep main.tex as the master file that only configures the paper and \input{}s section files.

Before drafting the front matter, re-read the one-sentence contribution from PAPER_PLAN.md. The Abstract and Introduction should make that takeaway obvious before the reader reaches the full method.

Section-Specific Guidelines

§0 Abstract:

• Use the 5-part flow from ../shared-references/writing-principles.md: what, why hard, how, evidence, strongest result
• 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:

• Follow the Introduction Blueprint exactly. Do not skip or reorder the narrative chain.
• Paragraph 1: Engineering demand — explain the real engineering or scientific need before introducing the algorithm.
• Paragraph 2: Problem emergence — derive the paper's problem from that demand and show why it is important and must be studied.
• Paragraph 3: Existing method categories — summarize the main categories of existing methods and what each category tries to solve.
• Paragraphs 4-6: 2-3 limitations — each limitation gets one paragraph. The first sentence must state the limitation clearly. The rest of that paragraph must explain only that limitation: why it happens, what it affects, and why existing methods struggle with it.
• Transition paragraph: one sentence in this form: "Based on the above, this paper proposes [algorithm/framework] to [solve P1-P3]."
• Contributions: write 3-4 contribution items. The first sentence of each item names the contribution. The next 3-4 sentences explain what algorithm/framework improvement was made and what effect or role it has. If experiments exist, the final contribution must report improvement over baseline algorithms.
• End with a brief section roadmap.
• 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 the existing method categories and limitations from the Introduction.
• Each category must answer: what this line of work solves, which locked problem or limitation it corresponds to, and where this paper improves a different link in the chain.
• 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
• Every key results paragraph must name the Problem ID or Contribution ID it supports, identify the table/figure/metric that provides evidence, and explain what the improvement over baselines means for the original engineering/scientific demand
• 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

Step 4.5: Theory Paper Consistency Pass (theory papers only)

Run this pass after drafting all sections and before building the bibliography.

Trigger it when PAPER_PLAN.md labels the paper as theory/analysis, or when the drafted sections contain five or more formal result environments (theorem, lemma, proposition, or corollary).

Proof source search: search the workspace for standalone full-proof sources whose names or contents indicate a canonical proof version (proof, appendix, full, complete, supplement, supplementary). If one exists, ask:

Inline full proofs from {file}? [Y/n]

Default to Y. If accepted:

• import the full theorem/lemma statement plus proof block into the appendix source;
• use the main-body theorem statement as the canonical public statement;
• do not leave placeholders such as "see supplementary proof document" or "proof omitted for brevity";
• preserve theorem labels, equation labels, and proof structure exactly;
• keep main-body proof sketches short, but never let the appendix be sketch-only when a full proof source exists.

Restatement audit: compare every theorem/lemma/proposition statement restated in the appendix against the main-body version. Audit statements, hypotheses, case splits, quantifiers, domains, notation, variable names, and terminology for defined objects. Resolve all mismatches before Step 5.

Step 5: Build Bibliography

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

1. Scan all \citep{} and \citet{} references in the drafted sections
2. Build a citation key list
3. 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
4. Write references.bib containing ONLY cited entries (no bloat)

Verified BibTeX Fetch (when DBLP_BIBTEX = true)

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

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

# 1. Search by title + first author
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)
# 3. Fetch real BibTeX
curl -s "https://dblp.org/rec/{key}.bib"

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

# If paper has a DOI or arXiv ID (arXiv DOI = 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 literature notes may mention papers from LLM memory — this fetch chain is the gate that prevents hallucinated citations from entering the final .bib.

If the DBLP/CrossRef flow is not enough, load ../shared-references/citation-discipline.md for stricter fallback rules before adding placeholders.

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

import re
# 1. Grep all \citep{...}, \citet{...}, and \cite{...} from all .tex files
# 2. Extract unique keys (handle multi-cite like \citep{a,b,c} or \cite{a,b,c})
# 3. Parse the full .bib file, keep only entries whose key is in the cited set
# 4. Write the filtered bib

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)

Step 6: Scientific Writing Quality Pass (5 audit passes)

After drafting all sections, run five sequential audit passes. De-AI polish is included as one part of this quality pass, not a replacement for it.

Pass 1: Clutter Extraction — strip sentences to their cleanest components, remove filler, and remove AI-isms.

Pass 2: Active Voice and Verb Vitality — identify who did what, convert unnecessary passive voice, and resurrect smothered verbs.

Pass 3: Sentence Architecture — flag sentences over 40 words, keep subject and verb close, put familiar context first and new information later, and ensure each paragraph does one job.

Pass 4: Keyword Consistency — apply the Banana Rule: do not rename defined technical terms just to avoid repetition. If Methods defines a group, variable, or technique name, Results, Discussion, tables, and captions must use the same term.

Pass 5: Numerical and Citation Integrity — check sample sizes, percentages, significant figures, figure/table values, and whether citations support the claims they are attached to.

Introduction Structure Check — verify before converting the Introduction draft to LaTeX:

• It starts from engineering/scientific demand, not from the algorithm.
• It derives the paper problem from that demand.
• It proves the problem is important and worth studying.
• It summarizes the main existing method categories.
• It states 2-3 limitations, each in a separate paragraph.
• Each limitation paragraph's first sentence states the limitation clearly.
• Each limitation paragraph explains only that limitation.
• It uses one sentence to introduce the proposed algorithm/framework based on the above.
• It contains 3-4 contribution items.
• Each contribution item starts by naming the contribution.
• Each contribution item explains the improvement mechanism and effect in 3-4 sentences.
• If experiments exist, the final contribution reports baseline improvement.
• It ends with section organization.

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

First apply the sentence-level clarity rules from ../shared-references/writing-principles.md:

• keep subject and verb close together
• put familiar context first and new information later
• place the most important information near the end of the sentence
• let each paragraph do one job
• use verbs for actions instead of nominalized nouns

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"

Step 7: Paper Architect/Reviewer Cross-Review

Delegate the complete draft to the paper architect/reviewer role:

Paper architect/reviewer task:
  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?
  8. Does the Introduction follow engineering demand -> problem importance -> existing methods -> 2-3 limitations -> proposed method -> 3-4 contributions -> organization?
  9. Do Related Work and Results Analysis explicitly align with the locked problems?

  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.

Step 8: Reverse Outline Test (from Research-Paper-Writing-Skills)

After drafting all sections:

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

Step 9: Final Checks

Before declaring done:

• All \ref{} and \label{} match (no undefined references)
• All citation commands (\citep{}/\citet{} for ML conferences, \cite{} for IEEE) 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)
• Complete Markdown drafts exist under paper/drafts/
• Every section has one Markdown draft and one matching LaTeX section file
• Introduction passes the Introduction Structure Check
• Related Work categories align with Introduction method categories and limitations
• Results/Analysis paragraphs map back to Problem IDs or Contribution IDs
• 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 \inputed by main.tex
• Section files match main.tex — file numbering and \input paths are consistent
• Venue-specific required sections/checklists satisfied (read ../shared-references/venue-checklists.md if needed)
• A skim reader can recover the main claim from the title, abstract, introduction, and Figure 1/captions

Key Rules

• Large file handling: If a large output cannot be written in one pass, use the available project-safe file editing mechanism to write it in smaller chunks and then verify the final file.

• Do NOT generate author names, emails, or affiliations — use anonymous block or placeholder

• Markdown first, LaTeX second — draft complete prose in paper/drafts/*.md, then write final content into paper/sections/*.tex

• Write complete sections, not outlines — both Markdown drafts and LaTeX sections should be complete, not placeholders

• One file per section — modular structure for easy editing

• One Markdown draft and one TeX file per section — keep filenames aligned across paper/drafts/ and paper/sections/

• Every claim must cite evidence — cross-reference the Claims-Evidence Matrix

• Problem alignment is mandatory — Introduction, Related Work, Method, and Results must all trace back to the locked 2-3 problems

• 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 — ML conferences (ICLR/NeurIPS/ICML) use natbib (\citep/\citet); IEEE venues use cite package (\cite{}, numeric). Never mix.

• Page limit rules differ by venue — ML conferences: main body to Conclusion, references/appendix NOT counted. IEEE: references ARE counted toward the page limit.

• Clean bib — references.bib must only contain entries that are actually \cited

• 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

• Front-load the contribution — do not hide the payoff until the experiments or appendix

Writing Quality Reference

• ../shared-references/writing-principles.md — story framing, abstract/introduction patterns, sentence-level clarity, reviewer reading order
• ../shared-references/venue-checklists.md — ICLR/NeurIPS/ICML/IEEE submission requirements to check before declaring done
• ../shared-references/citation-discipline.md — stricter fallback for ambiguous citations

Principles from Research-Paper-Writing-Skills:

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

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

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

Acknowledgements

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.