AI Research Reproducibility Copilot

Workflow Progress

Paper Uploaded
Hypotheses Extracted
Experiment Running
Report Generated

Settings

Configure experiments, data, execution, and reporting.

Experiment Settings

Defaults for running experiments and analyses.

Seed used to make runs deterministic across trials.

How many times to repeat the experiment for robust estimates.

Statistical confidence level used for intervals (e.g., 0.95).

95%

Primary metrics to compute and report.

Run experiments in an isolated environment with mocked side effects.