Under local magnetic Gauss constraints, vertex-disjoint regions can have commuting physical observable algebras that do not generate all observables on their union. We construct such counterexamples for every finite non-Abelian group on regular square tori, retaining the full magnetic physical space and original supports inside a fixed contractible disk. A peripheral projector selects inputs on which a joint transition preserves the constraints, while its loop factor alone fails the universal test over physical exterior completions. For a specified support family around a fixed eight-edge square loop, additivity fails if and only if the group is non-Abelian and each inner face is covered by at least one of its two prescribed peripheral edges. Leaving a face uncovered restores full additivity. For the same loop, additivity also holds with any region at vertex-set graph distance at least two. These results refute a recent group-double additivity proposition.
AI Summary
The paper asks whether commuting observable algebras of vertex-disjoint regions generate every observable on their union in quantum-double models subject only to magnetic Gauss constraints. Regional observables are defined using operators on the original edge supports that preserve the full magnetic physical space, followed by physical $C^*$-closure; no electric or global-sector projection is imposed.
For every finite non-Abelian group, it constructs counterexamples on square tori with $N,M\geq8$, with both supports contained in a contractible disk. For a fixed eight-edge square loop and a family of peripheral supports, additivity fails exactly when the group is non-Abelian and every inner face is covered by at least one of two prescribed contact edges. Thus 81 of 256 contact subsets fail; the minimum requires four contacts, one per face. An uncovered face restores equality for every finite group. Additivity also holds for this loop with any region avoiding a specified 32-edge neighborhood, including all regions at vertex-set graph distance at least two.
The proof distinguishes universal constraint preservation from preservation conditional on a peripheral projector. Isolated loop transitions permit only uniform central shifts; conditioning makes a noncentral transition legal. Center-orbit projectors separate this joint observable from the regional join by operator-norm distance exactly one. Face relabelings and matrix-unit factorization establish the equality cases. Supplementary results give complete algebra blocks for a wrapping pair and an independent $H_8$ counterexample.
These results contradict a stated group-double additivity proposition, but the coverage and buffer theorems concern the specified loop geometry, not arbitrary regions, electrically projected ground spaces, or continuum limits.
Little sign of AI-generated text
Written by gpt-6.1-sol from the paper's full text. Automated: it can be wrong.
Comments
24 pages, 3 figures; includes Supplemental Material
Under local magnetic Gauss constraints, vertex-disjoint regions can have commuting physical observable algebras that do not generate all observables on their union. We construct such counterexamples for every finite non-Abelian group on regular square tori, retaining the full magnetic physical space and original supports inside a fixed contractible disk. A peripheral projector selects inputs on which a joint transition preserves the constraints, while its loop factor alone fails the universal test over physical exterior completions. For a specified support family around a fixed eight-edge square loop, additivity fails if and only if the group is non-Abelian and each inner face is covered by at least one of its two prescribed peripheral edges. Leaving a face uncovered restores full additivity. For the same loop, additivity also holds with any region at vertex-set graph distance at least two. These results refute a recent group-double additivity proposition.
The paper asks whether commuting observable algebras of vertex-disjoint regions generate every observable on their union in quantum-double models subject only to magnetic Gauss constraints. Regional observables are defined using operators on the original edge supports that preserve the full magnetic physical space, followed by physical C∗-closure; no electric or global-sector projection is imposed.
For every finite non-Abelian group, it constructs counterexamples on square tori with N,M≥8, with both supports contained in a contractible disk. For a fixed eight-edge square loop and a family of peripheral supports, additivity fails exactly when the group is non-Abelian and every inner face is covered by at least one of two prescribed contact edges. Thus 81 of 256 contact subsets fail; the minimum requires four contacts, one per face. An uncovered face restores equality for every finite group. Additivity also holds for this loop with any region avoiding a specified 32-edge neighborhood, including all regions at vertex-set graph distance at least two.
The proof distinguishes universal constraint preservation from preservation conditional on a peripheral projector. Isolated loop transitions permit only uniform central shifts; conditioning makes a noncentral transition legal. Center-orbit projectors separate this joint observable from the regional join by operator-norm distance exactly one. Face relabelings and matrix-unit factorization establish the equality cases. Supplementary results give complete algebra blocks for a wrapping pair and an independent H8 counterexample.
These results contradict a stated group-double additivity proposition, but the coverage and buffer theorems concern the specified loop geometry, not arbitrary regions, electrically projected ground spaces, or continuum limits.
Earlier work on preferred decomposition for graph-local Hamiltonians established a layered picture rather than a single theorem: empirical structure selection under locality-sensitive selectors, identification of the correct quotient notion of uniqueness, local rigidity and local quantitative identifiability near exact realizations, stability under sufficiently small forbidden tails, and conditional global uniqueness once the exact zero set is known to be unique. The present paper shows that the unrestricted global step cannot be completed from graph locality alone. It gives an explicit connected counterexample family on star graphs: a commuting qubit Hamiltonian family preserved by an entangling star-CNOT circuit that exchanges leaf fields with edge Ising couplings while producing a genuinely inequivalent tensor-product decomposition. The same abstract Hamiltonian therefore has at least two exact admissible-support realizations, so exact zero-set identifiability fails on the physical quotient. The paper also shows why this does not contradict the earlier positive results. The counterexample lies entirely on a structurally degenerate locus excluded by the nondegeneracy hypotheses used in the local rigidity and exact reconstruction results: visible local directions are at most one-dimensional and the relevant complement algebras are abelian rather than full matrix algebras. The main foundational conclusion is sharp. Graph locality by itself does not determine a unique preferred decomposition even in the exact, static, connected setting. What remains open is a narrower problem: whether exact uniqueness holds on a suitably formulated nondegenerate locus that excludes commuting disentanglable obstruction classes of the type exhibited here.
We study algebraic locality principles on a 2+1D closed lattice in the presence of a Gauss law for a non-invertible symmetry. Prior work in arXiv:2509.03589 showed that when enforcing the Gauss law of an invertible symmetry, the principle of "Haag duality" is preserved exactly, and "disjoint additivity" is preserved after appropriate treatment of discreteness artifacts. Here we show that for a large class of non-invertible on-site symmetries, Haag duality is preserved exactly only for sufficiently nice, "cuspless" regions. For cusped regions, we instead have a weak form of Haag duality that requires adding a collar. Our results apply to double models with a purely magnetic constraint, and to the more general framework of constraints induced by the on-site action of a Hopf algebra. In particular, we treat a class of extended string-net models explicitly. We also demonstrate disjoint additivity for double models based on a group, and a weakened form of disjoint additivity in the setting of a general Hopf algebra.
Fix a quantum channel and two disjoint regions X, Y of its output. Operator-algebra quantum error correction assigns to each region a largest algebra of input observables it can exactly recover. We ask which pairs of such algebras occur. Our contribution is a realization theorem: for every pair of block-size vectors p, q and every nonnegative integer matrix m_ij with no zero row or column, an explicit channel has exactly those algebras as the maximal recoverable algebras of two disjoint outputs. Maximality is what makes this non-trivial: it forbids the channel from leaking anything else to either side. For a correctable-private pair the analogous realization is known (Kribs, Levick, Nelson, Pereira and Rahaman). The necessary direction is assembled from known material and we say so. Joint measurability forces the two algebras to commute. The decomposition of the code space into sectors C^{p_i} (x) C^{q_j} (x) C^{m_ij}, with the two algebras acting as diagonal copies and dim = sum_ij p_i q_j m_ij, is the Bratteli inclusion matrix of a commuting pair, which under complementary recovery is the one used in holographic quantum error correction. The identification of what both regions recover with the connected components of the bipartite graph of the nonzero m_ij is, at multiplicity one, a theorem of Vanrietvelde, Mestoudjian and Arrighi on partitions of C*-algebras; our residue is the extension to higher multiplicity, which their partitions exclude. Together: the quantum part of what two regions share is partitioned along the edges of a graph and never copied, the classical part surviving in both is exactly the connected components, and no admissible pattern is vacuous. Two disjoint heirs of a shared factor M_d inherit factors M_p, M_q only if pq divides d, sharpening pq <= d, so a shared qubit does not split. We also state the approximate counterpart, via the universal-cloning frontier.
We introduce double localization, a framework for proving spectral gaps of quantum Gibbs samplers through two complementary operations: geometric localization, which selects updates supported in small spatial regions, and interface localization, which focuses on a model-defined subspace of observables while remaining geometrically global. Our abstract gap theorem combines a global bound on this subspace, called the interface, with local quantum gap estimates under quantitative coupling and geometric assembly conditions. The resulting explicit lower bound controls the gap of the full dynamics without requiring the interface to be invariant under the generator. This separation allows global estimates, including classical comparison, to be combined with local control of the remaining quantum directions.
We apply the framework to Hamiltonians built from the vertex terms of Kitaev's finite-group quantum double construction, without plaquette terms, on finite simple three-regular graphs of girth at least six. For the specified local Davies dynamics, we prove an unconditional spectral-gap lower bound by a positive constant independent of graph size for every fixed nontrivial finite group and each fixed inverse temperature β with 0≤βJ<log(5/3), where J>0 is the coupling strength. For the smallest non-Abelian group S3, double localization yields a positive lower bound uniform in both graph size and the entire interval 0≤βJ≤log3, providing a guarantee that does not follow directly from existing results. This demonstrates how model-specific finite calculations can extend spectral-gap guarantees for the full quantum dynamics.
The Page–Wootters construction and the three equivalent pictures built on it all require a bipartition of a kinematical space into a clock and a system, with commuting Hamiltonians and no interaction term. In the literature that bipartition is input, and its non-interaction clause is acknowledged as an idealisation. We observe that a reductive dual pair supplies it as output: two subgroups of a symplectic group that are each other’s full centraliser act on the oscillator representation with a multiplicity-free joint decomposition, so the two factors commute by theorem rather than by assumption, the tensor structure is canonical rather than chosen, and the sector label is one both factors must agree on. Of six requirements of the relational construction, four are met exactly, one is reduced to a condition on the pairing bijection, and one — the covariant POVM that makes a clock — is not supplied at all. Three consequences. The statement that the clock–system tensor product does not survive to the physical Hilbert space becomes: the constraint selects the diagonal of the joint decomposition. Whether commuting subalgebras form an exact dual pair is finite and checkable, and in a related arena it was checked and answered negatively — a gauge algebra of dimension 12 with a double centraliser of dimension 82 — so exactness must be measured, not assumed. And the oscillator representation is a representation of the double cover, whose covering sign no observable of the pair can see. Finally, an obstruction of the form “one universal clock cannot supply a per-mode cost” is shown to move from universal to uniform without declaring it; a clock universal as an operator but non-uniform in its weights is a third option the standard enumeration does not contain. We name it and do not claim it works. Every object here is an algebraic object on a symplectic arena.
The toric code can be constructed as a gauge theory of finite groups on oriented two dimensional lattices. Here we construct analogous models with the gauge fields belonging to groupoids, which are categories where every morphism has an inverse. We show that a consistent system can be constructed for an arbitrary groupoid and analyze the simplest example that can be seen as the analog of the Abelian Z2 toric code. We find several exactly solvable models that have fracton-like features which include an extensive ground state degeneracy and excitations that are either immobile or have restricted mobility. Among the possibilities we study in detail the one where the ground state degeneracy scales as 2×2Nv, where Nv is the number of vertices in the lattice. The origin of this degeneracy can be traced to loop operators supported on both contractible and non-contractible loops. In particular, different non-contractible loops, along the same direction on a torus, result in different ground states. This is an exponential increase in the number of logical qubits that can be encoded in this code. Moreover the face excitations in this system are deconfined, free to move without an energy cost along certain directions of the lattice, whereas in certain other directions their movement incurs an energy cost. This places a restriction on the types of loop operators that contribute to the ground state degeneracy. The vertex excitations are immobile. The results are also extended to the groupoid analogs of Abelian ZN toric codes.