Skill UI

This guide details the best practices for packaging, documenting, and releasing research artifacts (code, models, data) for top-tier ML conferences. It covers two stages: preparing a non-identifiable archive for anonymous review, and managing the public release runway, focusing on reproducibility, licensing, and sustained maintenance.

A comprehensive guide for academic authors detailing how to establish and maintain rigorous reproducibility in Machine Learning research papers (e.g., ICLR). It instructs on mapping every central claim to verifiable evidence, including recording seeds, variance, specific compute budgets (training vs. inference), data preprocessing scripts, and ethical considerations. The statement must function as a public contract for the scientific community.

This guide assists researchers in determining the optimal venue for their theoretical machine learning contributions. It provides a framework for evaluating whether a result is best suited for COLT, or if it belongs in other top-tier conferences (e.g., NeurIPS, ICML) or journals (e.g., JMLR). It emphasizes differentiating between rigorous mathematical proofs (bounds, separations) and purely empirical findings.

This guide helps researchers determine if their Machine Learning project is suitable for ICLR, or if a different top-tier conference (such as NeurIPS, ICML, or CVPR) is a better fit. It emphasizes the need for deep representation-learning insights, mechanistic understanding, and theoretical contributions over mere application performance.

This guide helps transform technically sound machine learning drafts into highly readable, conference-ready papers suitable for ICLR and OpenReview. It emphasizes structuring the core learning-representation insight upfront, clarifying contribution types (method, theory, benchmark), optimizing for public skimming, and providing clear reviewer navigation paths. Use this when ensuring maximum impact and clarity in a challenging academic review environment.

A comprehensive guide for authors on drafting strategic, anonymous rebuttals and discussion replies for top-tier Machine Learning conferences like ICML. Learn how to triage reviewer objections by focusing on soundness, originality, significance, clarity, ethics, and reproducibility, adhering strictly to double-blind review constraints and maximizing impact in limited discussion rounds.

This guide provides a comprehensive framework for stress-testing Machine Learning experimental evidence before submission or rebuttal. It covers critical areas such as establishing strong baselines, conducting mechanism-isolating ablations, reporting variance, checking for data leakage, and disclosing compute costs. Use this to ensure your ML claims are scientifically sound, robust, and fair against rigorous academic scrutiny.

This guide provides structured methodologies for authors to position their research within the context of existing literature and concurrent submissions at top-tier machine learning conferences (e.g., ICML, NeurIPS). It emphasizes identifying the technical delta, addressing novelty concerns, and properly citing related work to maximize acceptance chances.

A comprehensive guide for researchers deciding the best venue for their machine learning manuscript. It helps evaluate whether a paper is a strong fit for ICML, or if it should be rerouted to specialized tracks, other top-tier conferences (NeurIPS, ICLR), or academic journals based on its core contribution type (theory, method, application) and novelty.

This comprehensive guide teaches researchers how to structure and refine machine learning papers for top-tier conferences like ICML. It focuses on achieving clarity, soundness, originality, and significance within strict page limits, providing actionable strategies to ensure the research withstands rigorous reviewer scrutiny.

A comprehensive guide for authors writing in computer vision, pattern recognition, or machine learning methods. This skill helps authors evaluate whether their manuscript meets the high methodological rigor, archival standards, and specific formatting requirements of IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI). It covers scope fit, evidence bar, structure, and pre-submission checklists.

This guide provides rigorous methodology for authors submitting security and machine learning research to top conferences. It emphasizes the need to 'freeze the world' of measurement by pinning target versions, microarchitectural states, and network conditions. It details how to report probabilistic evidence (like attack success rates) and manage ethical data release, ensuring findings remain valid and verifiable over time.