What problem does it solve? Writing a publishable ML/AI paper involves coordinating literature review, experiment design, statistical analysis, LaTeX drafting, citation verification, and venue-specific formatting — a process where hallucinated citations, missing baselines, and checklist violations cause desk rejections. This Skill provides a structured, iterative pipeline covering the full research lifecycle for NeurIPS, ICML, ICLR, ACL, AAAI, and COLM submissions. ## Core Features & Use Cases - Full Research Lifecycle: Eight phases from project setup and literature review through experiment execution, statistical analysis, drafting, self-review, and submission, with explicit feedback loops between phases. - Citation Verification Workflow: Mandatory 5-step programmatic citation process using Semantic Scholar, CrossRef, and arXiv APIs to eliminate hallucinated references, with a complete CitationManager implementation. - Venue-Specific Resources: Official LaTeX templates for six conferences, page limit references, and mandatory checklist documentation (NeurIPS 16-item checklist, ICLR LLM disclosure, ACL limitations section). - Experiment Infrastructure Patterns: Incremental saving for crash recovery, blind judge panel evaluation, statistical tests (McNemar's, bootstrap CIs, Cohen's h), human evaluation design, and publication-quality figure generation with colorblind-safe palettes. - Use Case: A researcher with experimental results asks the agent to draft an ICML submission — the Skill guides claim-to-experiment mapping, verified BibTeX generation, booktabs tables, and the reproducibility checklist before submission. ## Quick Start Ask the agent to help write a research paper from your existing experiment results, specifying your target venue such as NeurIPS or ICML.
