Context
A symbolic frontier is not yet a reduction. The concrete tuple must satisfy every finite-N field, support, reconstruction, interpolation, valuation, radius, parity, and gap-transfer obligation simultaneously.
Hypotheses and scope
- N >= 100, q is the least power of two at least N^242, K=N^34, and T=N^172.
- The two exceptional-loss classes receive separate q/4 budgets.
- The exact completeness estimate uses ell+1 <= N and r^2 <= 4R <= 4Nq.
Proof or evidence
The integer-arithmetic checker passes every listed side condition. The Lean development reaches the concrete Euclidean and two binary NP-hardness endpoints, with thirteen axiom probes reporting only standard Lean axioms and no admitted proof in the modified files.
Verification notes
The parameter audit, canonical checker output, complete constraint table, exact coefficient-16 margin, coefficient-four binary threshold, three final Lean endpoints, proof-hole scan, and pinned narrow-fork record were rechecked.
Limitations
- The large constants affect practicality even though the construction remains polynomial time.
- The executable end-to-end Lean reduction is instantiated at c=1/30, not uniformly at every c below 1/28.
- This retunes and closes the parameter bookkeeping of the retained reduction; it does not introduce a new lattice-hardness method or claim external review.
Notes
The immutable Boxlab fork commit 3ac608981eb48c432e77b42b899689260351ca7d compiles GapCVP.Comparator.gapCVP30IsNPHard and the two binary endpoints through the physical polynomial-time reduction. Thirteen principal axiom probes report only propext, Classical.choice, and Quot.sound. The exact proof uses separate quarter-field loss allocation plus ell+1 <= N to obtain the strict coefficient-16 margin and carries coefficient-four binary soundness through the explicit and physical interfaces.