Markdown Editor
Open the editor

AI Researcher Kit

Plan, run and publish ML research: preregistration, training plan, data management, experiments, benchmarks, human evaluation, conference paper and reproducibility.

Step 1

Planning

  1. Study Preregistration Template

    Standard

    Preregistration of hypotheses, design, sampling plan and stopping rule, variables, analysis plan, exclusions and deviations log

    OSF preregistration (concepts)

  2. Model Training Plan Template

    Full

    Objective and success criteria, data mix and splits, baseline, architecture, 6ND compute budget in KaTeX, ablations, checkpoints, stop rules, gates, risks

    Planning concepts from Kaplan et al. 2020 and Hoffmann et al. 2022

  3. Data Management Plan (DMP) Template

    Full

    Funder-style DMP – storage, sharing, privacy

    NIH/NSF/EU DMP conventions
Step 2

Evaluation and safety

  1. ML Experiment Tracking Report Template

    Full

    Hypothesis and decision rule, setup, config and seeds, data versions, results vs baseline with variance, paired test, ablations, failure analysis, decision

    Reporting informed by the NeurIPS Paper Checklist (2026)

  2. Model Benchmark Report Template

    Full

    Benchmark choice and validity, contamination checks, protocol, prompts and decoding, repeated runs, confidence intervals, McNemar test, comparison table

    Practice from HELM (TMLR 2023), Miller 2024 and Sainz et al. 2023

  3. Human Evaluation Protocol Template

    Full

    Task and rubric, rater recruitment and training, ethics and consent, sample size, blinding, Cohen's/Fleiss' kappa, Krippendorff's alpha, QC, analysis

    Human Evaluation Datasheet (HumEval 2022); van der Lee et al. 2021
Step 3

Documentation

  1. ML Conference Paper Template (NeurIPS Checklist)

    Full

    ML conference paper: abstract, method, experiments, results, limitations, broader impact, reproducibility statement, appendix, NeurIPS 2026 checklist, ACL note

    NeurIPS 2026 Paper Checklist (16 items, paraphrased); ARR Responsible NLP checklist

  2. ML Reproducibility Checklist Template (REFORMS)

    Full

    REFORMS-aligned ML science checklist: 32 items in 8 modules, leakage audit worksheet, evaluation protocol, code and data release, compute reporting

    REFORMS (Kapoor et al., Science Advances 2024), 32 items in 8 modules, paraphrased

  3. Model Card Template (Mitchell et al. 2019)

    Full

    Model details, intended and out-of-scope uses, factors, metrics, data, disaggregated results, ethics, caveats

    Model Cards for Model Reporting (Mitchell et al., 2019)

  4. Datasheet for Datasets Template (Gebru et al. 2021)

    Full

    Dataset motivation, composition, collection, labelling, uses, distribution, maintenance, privacy and consent

    Datasheets for Datasets (Gebru et al., 2021)

  5. Peer Review Report Template

    Standard

    Peer review report: summary, assessment, numbered major and minor comments, methods and ethics checks, recommendation, editor notes

    COPE Ethical Guidelines for Peer Reviewers (cited)