Stats
Actions
Tags
From neurips-skills
Audits NeurIPS experimental evidence: baselines, ablations, robustness, compute, data splits, negative results, and claim calibration. Prepares rebuttal-ready clarifications.
How this skill is triggered — by the user, by Claude, or both
Slash command
/neurips-skills:neurips-experimentsThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Use this skill before submission or rebuttal when the main question is whether the evidence supports
Use this skill before submission or rebuttal when the main question is whether the evidence supports the NeurIPS claim. It is not enough to win a leaderboard; reviewers need to know why the result is scientifically meaningful.
Prepare small, high-signal clarifications that can fit in an author response: a missing baseline table, a sanity check, an error analysis, a variance estimate, or a concise proof sketch. Do not depend on a complete post-review paper rewrite.
[Evidence status] strong / adequate / weak
[Main unsupported claim] <claim>
[Critical missing experiment] <baseline/ablation/robustness/data/compute>
[Small rebuttal result] <result feasible during response>
[Claim rewrite] <narrower claim if evidence stays as is>
npx claudepluginhub brycewang-stanford/awesome-journal-skills --plugin neurips-skillsAssists with NeurIPS reproducibility: aligns Paper Checklist with the paper, writes code/data instructions, sets seed/compute disclosure, and decides MLRC vs. main track.
Audits ML experiments for ICML submission/rebuttal: baselines, ablations, variance, data leakage, compute disclosure, reproducibility, negative results.
Audits IJCAI/IJCAI-ECAI experiments for baselines, ablations, statistical evidence, hyperparameters, compute, dataset handling, ethics, and reproducibility.