A preprint posted to arXiv — not yet peer-reviewed — claims to establish a uniform spectral gap for the spin-3/2 AKLT model on the hexagonal lattice, a long-standing open problem in quantum many-body physics. The authors develop a general sufficient condition under which overlapping projected entangled-pair state (PEPS) boundary states satisfy an "approximate factorization" criterion, a technical property that, once verified, implies a nonzero gap above the ground state of the associated parent Hamiltonian.
The central construction decomposes the boundary response generated by cutting bonds through a rectangular region into a "common" contribution shared across four associated sub-regions and a remainder term. A single linear transport map is shown to absorb the common part and produce compatible factorization operators; the error in the factorization is then controlled entirely by the remainder, which the authors bound using a local comparison of boundary states together with a scalar Kotecký–Preiss cluster-expansion estimate. For Ising PEPS, the analogous bound reduces to a Dobrushin–Shlosman-type condition familiar from classical statistical mechanics. In both cases the factorization error decays exponentially in the overlap width, up to a prefactor linear in the cut length.
If the analysis withstands peer review, the result would provide the first rigorous spectral-gap certificate for the hexagonal AKLT model and, more broadly, a systematic pathway from the locality of PEPS boundary correlations to gap proofs for two-dimensional parent Hamiltonians. Readers should note that the work is a preprint and the technical claims — including the Kotecký–Preiss estimate for AKLT loop expansions and the physical projector bound — have not yet been independently verified.