What problem does it solve?
Provide a structured, auditable pre-submission quality-assurance and rebuttal workflow for near-final machine learning conference papers, reducing desk-reject risks and improving clarity, reproducibility, and citation integrity.
Core Features & Use Cases
- Two-view review: section-by-section edits plus a consolidated prioritized P0/P1/P2 issue list with verification notes and recommended fixes.
- Rebuttal & response drafting: parse reviewer comments, classify issues, choose strategies, and produce point-by-point rebuttal drafts and a minimal revision plan.
- Citation integrity & LaTeX safety: explicit no-hallucination rules, citation audits, and edits that preserve LaTeX semantics (\cite{}, math, labels).
- Execution modes: targeted runs for relevant tracks or full-parallel mode to run all audit tracks independently and synthesize results.
- Use Case: final QA before ICML/ICLR/NeurIPS/AAAI submission, camera-ready checking, or drafting a structured rebuttal to reviewer comments.
Quick Start
Run a section-by-section pre-submission review on my near-final ML paper draft and produce a consolidated P0/P1/P2 action list plus a draft point-by-point rebuttal.