ml-paper-writing

Draft ML/AI conference papers from research repositories and verify citations.

Updated Apr 10, 2026
One-click install
npx skills add https://github.com/KarlinskyS/hermesSkills --skill ml-paper-writing-karlinskys
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ml-paper-writing
Source: https://github.com/KarlinskyS/hermesSkills/tree/main/research/ml-paper-writing
Command: npx skills add https://github.com/KarlinskyS/hermesSkills --skill ml-paper-writing-karlinskys

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires semanticscholar, arxiv, habanero, requests, and includes references (resource) components.

What problem does it solve?

Research teams and individual scientists often struggle to turn code, results, and experimental artifacts into a coherent, submission-ready ML/AI paper for top venues. This Skill reduces the burden of structuring arguments, producing drafts, verifying citations, and formatting for conference templates so teams can focus on science rather than submission mechanics.

Core Features & Use Cases

  • Repository-to-Draft Workflow: Explore a research repo, identify main contributions, and produce a full first draft (abstract, introduction, methods, experiments, related work, limitations).
  • Citation Verification: Search literature via Semantic Scholar/arXiv/CrossRef, fetch BibTeX programmatically, and mark unverifiable references explicitly as placeholders.
  • LaTeX Templates & Format Conversion: Provide and verify official LaTeX templates (NeurIPS, ICML, ICLR, ACL, AAAI, COLM) and convert papers between venue formats while preserving checklists and required sections.
  • Reviewer & Checklist Guidance: Include venue-specific checklists, reviewer-expectation heuristics, figure and writing best practices, and reproducibility requirements.
  • Use Case Example: Starting from an experiments repo, produce a NeurIPS-style draft, verify every citation via APIs, generate a .bib, and prepare a camera-ready PDF with the correct template.

Quick Start

Draft a complete NeurIPS-style first draft from this repository, verify all citations programmatically, and prepare a LaTeX submission using the neurips2025 template.

Frequently Asked Questions about ml-paper-writing

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

FAQPage Schema
How do I write an ML conference paper directly from my code repository?

You can generate a complete first draft of an ML conference paper from a research repository by analyzing experimental artifacts, extracting main contributions, and automatically drafting sections like abstract, introduction, methods, and limitations.

How do I verify citations and fetch BibTeX for an academic writing paper?

Citation verification is handled by querying Semantic Scholar, arXiv, and CrossRef APIs to fetch BibTeX entries programmatically, ensuring references are validated and unverifiable citations are explicitly marked as placeholders.

Can I convert my academic paper between NeurIPS, ICML, and ICLR LaTeX templates?

Yes, the Skill provides official LaTeX templates for NeurIPS, ICML, ICLR, ACL, AAAI, and COLM, and can convert papers between these venue formats while preserving required checklists and sections.

Does this academic writing workflow include conference submission checklists and reviewer guidelines?

Yes, the workflow integrates venue-specific checklists, reviewer-expectation heuristics, figure best practices, and reproducibility requirements to ensure conference submissions meet reviewer standards.

What happens to unverifiable citations when drafting ML paper writing submissions?

During ML paper writing, unverifiable citations are explicitly marked as placeholders, ensuring transparency in the draft while you locate the missing reference data before final submission.

Do I need arXiv and Semantic Scholar dependencies to generate a literature review for my paper?

Yes, the Skill relies on arXiv, Semantic Scholar, CrossRef, and Habanero dependencies to search literature, fetch BibTeX entries, and verify citations during the automated literature review process.