What problem does it solve? Drafting ML conference papers from research repositories is slow and error-prone, especially structuring the narrative, meeting venue-specific formatting rules, and avoiding hallucinated citations that can cause desk rejection. ## Core Features & Use Cases - End-to-End Paper Drafting: Explores a research repository, identifies the contribution, and produces complete section-by-section drafts for NeurIPS, ICML, ICLR, ACL, AAAI, and COLM. - Citation Verification Workflow: Searches Semantic Scholar and arXiv, fetches BibTeX programmatically via DOI, and marks unverifiable references as explicit placeholders instead of inventing them. - Conference Templates & Conversion: Provides LaTeX templates per venue, page-limit guidance, checklists, and workflows for converting a rejected submission to another conference format. - Use Case: Given a repo with experiment results, generate a full NeurIPS draft with verified related-work citations, then convert it to ICML format after rejection. ## Quick Start Use the ml-paper-writing skill to draft a NeurIPS paper from this research repository and verify all citations programmatically.
