Datasets:

---

FINAL-Bench

/

OSC-Superconductor

Like 39

Follow

FINAL_Bench 376

Thank you — a number-projected d-wave BCS/AGP + Jastrow approach is a serious one. Your claimed A_d=0.0727 exceeds our 6×6 value (0.056); for fairness every entry is currently scored on the material's standard value, with your claim kept as a “claimed” value. Please share reproducible code under the common conditions (pair operator, periodic BC, max-distance, U/t=8; a 4×4 value helps) and we will reproduce it — a verified higher A_d may exceed the baseline.

On the score change you noticed: every entry is now scored on the material's standard value (OSC-01613 → 22.01), and your claimed A_d = 0.0727 is kept as a “claimed” value, not lost. Your number-projected d-wave BCS/AGP + Jastrow is a serious method — please share reproducible code under the common conditions (a 4×4 value helps) and we will reproduce it; a verified higher A_d may exceed the baseline.

Thank you for the clarification.
I have prepared and sent the full reproducibility package for my OSC-01613 / Br2Cu Gold claim by email.
Conservative submitted result:

• A_d = 0.072749464952
• E = -20.688132457584
• projected score = 28.7546625
  The package includes the frozen checkpoint, exact-core reproduction code, independent-training-seed robustness checks, SHA256 manifests, and the public 4×4 reference together with my 4×4 operator audit.
  One caveat is documented explicitly: my open reconstruction of the 4×4 operator does not exactly reproduce all published 4×4 reference values, so I did not rescale or modify the observable to force agreement.
  The sealed reproduction ZIP passed a fresh-extraction test before sending.

Thank you — we received your reproducibility package. We will run the exact-core reproduction and, if the A_d holds under our independent verification, mark it method_verified (it may then exceed the baseline). We appreciate the explicitly documented 4×4 operator caveat.