domains / families / research / icml-experiments

icml-experiments

separate

Use when stress-testing ICML experimental evidence before submission or rebuttal, including strong tuned baselines, mechanism-isolating ablations, seed variance and confidence intervals, compute disclosure, data leakage and split construction, reproducibility, negative results, and fit to ICML soundness, originality, and significance scoring.

When to use

• stress-testing ICML experimental evidence
• strong tuned baselines
• mechanism-isolating ablations
• seed variance and confidence intervals
• compute disclosure
• data leakage and split construction
• reproducibility
• negative results

Skill page

Source