Answers are grounded in this paper and cite their sources.
Preprint and verification data/scripts for the paper "Edge-pair contraction for O(4) spins and strong coupling in four-dimensional SU(2) lattice Yang-Mills theory". We prove a contraction criterion for O(4) spin systems that updates the two spins of an edge jointly. On every graph of maximal degree at most Δ, with arbitrary orthogonal edge transports, it gives a unique Gibbs measure, exponential decay of covariances of Lipschitz observables uniformly in the volume and the boundary condition, and Wasserstein contraction of a heat-bath dynamics, for all couplings β < b_Δ, where b_Δ is an explicit threshold. In exact arithmetic b_4 > 1.0175, b_6 > 0.672488 and b_8 > 0.50258, beyond 4/Δ, the exact threshold of the worst-case single-site criterion in the chordal metric. For four-dimensional SU(2) lattice Yang–Mills theory with Wilson action, normalized so that the plaquette weight is exp((β/2) tr U_p), we apply the criterion twice. With a cluster expansion of the remaining links in a line gauge it gives, for every 0 < β ≤ 0.672487, a unique infinite-volume Gibbs measure, exponential clustering of gauge-invariant local observables, and a spectral gap of the transfer matrix in the physical Osterwalder–Schrader space; the largest explicit window for these statements known to us was β ≤ 0.51. On slabs it gives Wilson's area law |⟨W_ℓ⟩| ≤ exp(−c R T) with prefactor one for every rectangular loop and every 0 ≤ β < b_6, past the previous bound 2/3, for every Gibbs measure and, on the torus, for loops with sides at most half the period; the same argument gives the area law in three dimensions for every β < b_4. After explicit analytic tail bounds, every finite inequality is decided in exact rational arithmetic; the programs are supplied. Files: the paper (PDF); YM_certificates.zip with folders C1-C5 holding the verification programs, their inputs and outputs, one folder per group of statements, together with README.md, which maps every statement to its program, output and command, and SHA256SUMS.txt, the checksums of all files in the archive; a copy of README.md is also given separately.
No takes yet. Share an insight, caveat, or question.
Yancun Yan (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: