REVIEW PREPARATION ARTIFACT — LOCAL DRAFT

This ZIP is a local preparation copy. It has not been externally submitted,
distributed, peer reviewed, or accepted. The submitted manuscript is supplied
separately; original manuscript source/PDF are intentionally omitted here.

Included: implementation, configurations where separate files exist, tests,
dependency locks, protocol and claim-to-evidence documentation, full and pilot
results, verification records, and previously generated tables/figures.
Git metadata, CI workflows, software caches, and unrelated or untracked files
are excluded. All numerical results, source code, protocols, and test logic
are preserved without change. No new experiments were run while packaging.
Generated table fragments remain under paper/generated/ because the analysis
scripts use that output directory; they are not a manuscript submission.

CONTENTS.json lists files in this review copy. Detailed original/export hashes
and copy transformations are retained outside this ZIP in local audit records.
Original experiment manifests retain their scientific provenance checksums;
those checksums refer to the unchanged experiment sources and result data.

ANONYMOUS REVIEW DRAFT
The original project's copyright-holder identity has been redacted only in
this local LICENSE copy; all other MIT terms are preserved. The original
LICENSE remains unchanged. This temporary reviewer copy requires author
review before any external distribution. No third-party notice was changed.
The README's author name is withheld; its material AI-assistance disclosure
and scientific limits remain. A named original paper is not included.

REPRODUCTION
Use Python 3.12; the recorded environment used Python 3.12.13, NumPy 2.5.3,
and Matplotlib 3.11.2. No external dataset or model API is needed.
From this directory, create/activate an environment and run:
  python -m pip install -r requirements-lock.txt
  python -m pytest -q
  python -m src.verify --reference results/full
  python -m src.run --config configs/smoke.json --output reproduction/smoke
  python -m src.run --config configs/full.json --output reproduction/full
  python -m src.verify --reference results/full --reproduction reproduction/full
  python -m src.analyze --results reproduction/full --figures reproduction/figures --tables reproduction/tables

Preserved full evidence: 96 exact evaluations, 800 sampled fits, and 3 controls.
Pilot evidence is separately retained. The fitting/verification source code
is byte-identical to the original tracked version. source_hashes and artifact
checks in original run manifests remain usable; no fitted values, seeds,
metrics, timings, or claimed outcomes were altered to anonymize this copy.
The original clean-verification record contains a relative staging path,
not an absolute personal path; that record is retained unchanged.

Scientific scope: exact additive/pairwise synthetic reward-projection
diagnostics, not a neural MARL or original LOMAQ reproduction. Identity scans
cover explicit names, the user account, absolute personal paths, and available
image/PDF metadata; they do not guarantee anonymity against linking research
content with information outside the submission.
