skills.rest
superfhp
superfhpCOMMUNITY
Explore all skills by this author →
GitHub

research-paper-writing

Guides end-to-end ML research paper production from experiment design through conference submission.

Updated May 13, 2026
Search Tags:#citations#statistical analysis#latex#research paper#experiments#neurips#conference submission
One-click install
npx skills add https://github.com/superfhp/lumi-agent-body --skill research-paper-writing-superfhp
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: research-paper-writing
Source: https://github.com/superfhp/lumi-agent-body/tree/main/skills/research/research-paper-writing
Command: npx skills add https://github.com/superfhp/lumi-agent-body --skill research-paper-writing-superfhp

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires semanticscholar, arxiv, habanero, requests, scipy, numpy, matplotlib, SciencePlots, and includes references (resource) and assets (resource) components.

What problem does it solve? Writing a publication-ready ML/AI paper involves coordinating literature review, experiment design, statistical analysis, LaTeX drafting, citation verification, and venue-specific submission requirements. This Skill provides a structured pipeline covering the full research lifecycle for NeurIPS, ICML, ICLR, ACL, AAAI, and COLM submissions. ## Core Features & Use Cases - Full Research Pipeline: Eight phases from project setup and literature review through experiment execution, analysis, drafting, self-review, and submission, with explicit feedback loops. - Citation Verification Workflow: Programmatic BibTeX retrieval via Semantic Scholar, CrossRef, and arXiv APIs to prevent hallucinated citations, with a mandatory 5-step verification process. - Experiment & Analysis Patterns: Incremental result saving, crash recovery, blind judge panels, McNemar's test, bootstrapped confidence intervals, and publication-quality figures with colorblind-safe palettes. - Venue Templates & Checklists: Official LaTeX templates for six conferences plus mandatory checklist requirements (NeurIPS 16-item checklist, ACL Limitations section, ICLR LLM disclosure). - Use Case: A researcher with experimental results asks the agent to draft an ICML submission. The Skill verifies all citations programmatically, generates figures with error bars, applies the ICML 2026 template, and runs a simulated reviewer pass before submission. ## Quick Start Use the research-paper-writing skill to draft a NeurIPS paper from the experiment results in my results/ directory, starting with a literature review and verified citations.

Frequently Asked Questions about research-paper-writing

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I write a machine learning paper for NeurIPS or ICML?▼

Follow a phased pipeline: define a one-sentence contribution, map every experiment to a claim, run baselines with statistical tests, then draft using the venue's official LaTeX template. Complete the mandatory checklist (NeurIPS requires 16 items) before submission to avoid desk rejection.

How do I prevent hallucinated citations in AI-generated papers?▼

Never generate BibTeX from memory. Search Semantic Scholar for the paper, verify it exists in at least two sources, retrieve BibTeX via DOI content negotiation from CrossRef, and confirm the cited claim appears in the abstract. Mark unverifiable entries as placeholders.

What statistical tests should I report in an ML paper?▼

Report error bars (specifying standard deviation or standard error), 95% confidence intervals via bootstrapping, and McNemar's test for pairwise method comparisons. Include effect sizes like Cohen's h and state the number of runs and sample sizes.

Does ICLR require disclosing LLM use in paper writing?▼

ICLR requires disclosure only if LLMs played a significant role in research ideation or writing such that they could be considered contributors, described in a separate appendix section. Grammar checking and minor editing assistance do not require disclosure.

When should I use human evaluation instead of automated metrics?▼

Use human evaluation for subjective generation tasks like summarization, dialogue, and open-ended text where automated metrics miss fluency or helpfulness. Report annotator counts, inter-annotator agreement (Krippendorff's alpha), compensation, and use pairwise comparisons for reliability.

What are the page limits for major ML conferences?▼

NeurIPS allows 9 pages, ICML 8 pages, ICLR 9 pages, ACL 8 pages for long papers, AAAI 7 pages, and COLM 9 pages for main content. References and appendices are unlimited at all six venues, with camera-ready versions typically gaining one page.

Related Skills

Explore More Education & Research Skills→
ToqsickToqsick

arxiv

Search arXiv by keywords or IDs and retrieve abstracts and full papers.

Community
Advanced
EmaRimoldiEmaRimoldi

post-acceptance

Plan and produce conference talks, posters, and promotional content.

Community
Advanced
dogekikidogekiki

paper-claim-audit

Compare quantitative claims in research papers against raw JSON, CSV, and TSV result files.

Community
Advanced
llm011llm011
9

research-paper-writing

Coordinate ML research paper writing from literature review to submission preparation.

Official
Advanced