ML Paper Writing for Top AI Conferences

Expert-level guidance for writing publication-ready papers targeting NeurIPS, ICML, ICLR, ACL, AAAI, and COLM. This skill combines writing philosophy from top researchers (Nanda, Farquhar, Karpathy, Lipton, Steinhardt) with practical tools: LaTeX templates, citation verification APIs, and conference checklists.

Default operating order

Use this skill in the following order unless the task is unusually narrow:

1. lock the operating mode from references/OPERATING-MODES.md,
2. understand the repo or draft context,
3. use references/citation-workflow.md as the canonical citation authority,
4. load venue- or template-specific references only after the main writing path is clear.

Google Scholar may still help with manual discovery, but it is not the canonical verification authority in this skill. Default verification should use programmatic sources such as Semantic Scholar, CrossRef, and arXiv.

Claim ledger gate

Before a project plan, experiment note, or literature summary becomes manuscript prose:

• identify the Claim Candidate or Evidence Record that supports the sentence,
• preserve allowed wording and forbidden stronger wording,
• keep project plans as hypotheses unless experiment artifacts or verified papers support them,
• do not turn related-work motivation into evidence for the paper's own result,
• mark unsupported claims as [CLAIM NEEDS EVIDENCE] instead of polishing them.

If the repo context is clear enough for a first draft, still apply this gate before stating contributions, results, related-work contrasts, or rebuttal-facing claims.

Core Philosophy: Collaborative Writing

Paper writing is collaborative, but Claude should be proactive in delivering drafts.

The typical workflow starts with a research repository containing code, results, and experimental artifacts. Claude's role is to: