research-paper-writing

Write ML research papers for NeurIPS, ICML, and ICLR from experiment design through submission.

Updated Aug 22, 2026
One-click install
npx skills add https://github.com/vivekgoquest/hermes-agent-stable --skill research-paper-writing-vivekgoquest
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: research-paper-writing
Source: https://github.com/vivekgoquest/hermes-agent-stable/tree/main/skills/research/research-paper-writing
Command: npx skills add https://github.com/vivekgoquest/hermes-agent-stable --skill research-paper-writing-vivekgoquest

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? Producing a publication-ready ML/AI paper requires coordinating literature review, experiment design, statistical analysis, LaTeX writing, citation verification, and venue-specific formatting — a process where missed baselines, hallucinated citations, or formatting errors cause desk rejection. ## Core Features & Use Cases - End-to-End Pipeline: Covers project setup, iterative literature search, claim-to-experiment mapping, execution monitoring with cron patterns, statistical analysis (McNemar's test, bootstrapped CIs), drafting, self-review, and submission. - Citation Verification: Fetches BibTeX programmatically via Semantic Scholar, CrossRef, and arXiv APIs with a mandatory 5-step verification workflow to prevent hallucinated references. - Venue Templates & Checklists: Ships official LaTeX templates for ICML 2026, ICLR 2026, NeurIPS 2025, ACL, AAAI 2026, and COLM 2025, plus mandatory checklist requirements per venue. - Use Case: Given an existing codebase with experiment results, design the remaining experiments, run them with monitoring, compute significance tests, and produce a complete ICML-formatted draft with verified citations. ## Quick Start Use the research-paper-writing skill to turn my experiment results in this repository into a full ICML submission draft with verified citations and statistical analysis.

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 an ML paper for NeurIPS or ICML with an AI agent?▼

Follow the phased pipeline: set up the workspace and contribution statement, run an iterative literature search, map claims to experiments, execute with monitoring, analyze results with significance tests, then draft using the venue's official LaTeX template. The process is iterative, with reviews feeding back into new experiments.

How to prevent hallucinated citations in AI-generated papers?▼

Never generate BibTeX from memory; fetch it programmatically via DOI content negotiation from CrossRef. Verify each paper exists in at least two sources such as Semantic Scholar and arXiv, and mark anything unverifiable as [CITATION NEEDED] for manual review.

What statistical tests should ML papers report?▼

Report error bars with the method specified, 95% confidence intervals via bootstrapping, McNemar's test for paired method comparisons, and effect sizes such as Cohen's h. Always state sample sizes and number of independent runs.

Which LaTeX templates are included for conference submissions?▼

Official templates are provided for ICML 2026, ICLR 2026, NeurIPS 2025, ACL, AAAI 2026, and COLM 2025, each with the correct style file, bibliography style, and example document. Page limits and anonymization rules are documented per venue.

When should I use autoreason versus critique-and-revise for paper refinement?▼

Use autoreason for mid-tier models and constrained tasks where the generation-evaluation gap is large. Use critique-and-revise or single pass for frontier models on unconstrained tasks, since autoreason suffers synthesis drift without scope constraints.

Does this pipeline support human evaluation for NLP papers?▼

Yes, it covers annotator selection, pairwise versus Likert scales, inter-annotator agreement with Krippendorff's alpha or Cohen's kappa, crowdsourcing platform choice, compensation reporting, and IRB guidance required by ACL venues.