Use cases

Verify a benchmark or paper

Independently check that a repo's reported numbers reproduce — and aren't contaminated.

The core job: point Calma at a benchmark, eval, or paper repository and get an independent, third-party verdict on whether its reported numbers actually hold up. No self-grading — a deterministic re-run and recompute, neutral by construction.

This is the workflow for benchmark maintainers, eval vendors, artifact-evaluation committees, and anyone who needs to trust a number before they publish or buy on it.

What Calma checks

QuestionHowVerdict
Does the reported metric reproduce?Re-run the repo, recompute the metric from the raw outputs, diff itCONFIRMED / REFUTED
Do the committed result files regenerate?Re-run the repro script, diff regenerated vs committed cellsreproduced / not-reproduced
Is the benchmark contaminated?Train/test overlap (verbatim + near-duplicate) detectionINVALIDATED on leakage

Run it

# recompute the reported metric + check the committed result files regenerate
calma verify owner/benchmark-repo --deep --artifact-repro
  • --deep clones into a network-isolated sandbox, re-runs the repo, and recomputes each reported number.
  • --artifact-repro additionally diffs the repo's committed result files (leaderboard CSVs, results.json, per-run metrics) against what a fresh run regenerates — cell by cell. A number that doesn't come back is flagged not-reproduced with the exact file and cell.

If the repo's run command needs a config or data argument, name it:

calma verify owner/repo --deep --artifact-repro --entry "reproduce.py --config configs/main.yaml"

Two verifiable shapes

Calma lands a clean verdict when a repo has a runnable entrypoint plus one of:

  1. Committed predictions + labels + a reported score → Calma recomputes the standard metric and returns CONFIRMED / REFUTED. (Live leaderboards usually hide labels; reproducibility artifacts commit both.)
  2. A repro script + committed result files → artifact-reproduction diffs every regenerated cell against the committed one.

Why the verdict is trustworthy

The verdict is a single deterministic function of the recompute — it is fail-closed and can never be a false confirm. When Calma can't bind a number to a computation, it says INCONCLUSIVE rather than guessing. That is the whole point of a neutral auditor: a CONFIRMED means the number reproduced, full stop.

Gate it

calma verify owner/repo --deep   # exits non-zero on REFUTED / INVALIDATED

Drop that into CI (or the GitHub Action / PR check) so a regression that breaks a reported number fails the build.