Skip to content

Latest commit

 

History

History
25 lines (15 loc) · 1.09 KB

File metadata and controls

25 lines (15 loc) · 1.09 KB

Baseline Comparison

Baseline target

  • Classical theorem statements, topic-local proof scripts, and internal diagnostic outputs.

Current comparator package

  • Comparator or reference scripts should be taken from topic-local Code/04_Competitor/ when present.
  • If no dedicated competitor script exists, the baseline is the cited source dataset or reference model listed in DATA_MANIFEST.md.

Comparison metrics

  • consistency checks, bounded-domain counterexample searches, and script-reported scaling or stability diagnostics

Acceptance boundary

  • This file does not certify a final pass/fail result.
  • Until the benchmark is rerun with a saved artifact, comparison language must remain internal benchmark comparison.
  • A future hardening pass must record the exact numeric threshold, generated artifact, timestamp, environment, and dataset hash.

Claim boundary

  • This baseline comparison can support only conservative wording such as matched selected benchmarks or internal comparison workflow.
  • It does not support wording such as solved, verified, exact, unified, or production grade.