What problem does it solve? Drafting and revising academic papers for ML/CV/NLP venues often suffers from unclear paragraph flow, unsupported claims, weak introductions, and reviewer-facing presentation issues. This Skill provides section-by-section writing guidance, templates, and an adversarial review checklist to produce reviewer-friendly drafts. ## Core Features & Use Cases - Section-Specific Guides: Dedicated references for Abstract, Introduction, Related Work, Method, Experiments, and Conclusion, loaded only when needed. - Claim-Evidence Alignment: Enforces that every major claim in the Abstract and Introduction is backed by experimental evidence, with a claim-evidence map output. - Adversarial Self-Review: A five-dimension checklist (contribution, writing clarity, experimental strength, evaluation completeness, method design soundness) to catch rejection risks before submission. - Use Case: When revising an Introduction, the Skill applies templates for task framing, technical-challenge chains, and pipeline presentation, then runs reverse outlining to verify paragraph flow. ## Quick Start Use the research-paper-writing skill to rewrite my paper's Introduction section and check that every claim is supported by my experiments.
