{"id":"ec39b1f9-79bd-4db6-ae48-bf13a3ce360a","arxiv_id":"2510.21399","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"A modified Villain action built from heat kernels on image subgroups defines projective-limit measures whose infinite-lattice limit is translation-invariant and massless at all couplings.","lead":"This paper constructs continuum and infinite-volume limits of Abelian polyhedral gauge theories using projective systems of heat-kernel measures. The resulting infinite-lattice model is translation-invariant and massless for every coupling, unlike the standard U(1) lattice gauge theory.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The claimed contraction property for subdivision maps in §5.4 is false for d>2, so the Euclidean-invariant continuum limit (5.10) is not a projective system in LG; Theorem 5.7 is unaffected.","rationale":"We independently checked the reader's weakest assumption and concur. The contraction claim in §5.4 is not merely unproved but false: an explicit exact 2-cochain exhibits norm expansion for all d>2. Consequently the projective limit in LG over the subdivision poset Q_d is ill-defined, so the Euclidean-invariant continuum-limit measure μ_{β,∞} is not constructed. The central theorem of the paper, Theorem 5.7, however, is independent of §5.4: its proof via Lemma 5.6 uses only the Fourier analysis of the projection Π on ℓ^2(Z^d), and we found no error in that calculation (the decay n^{-d} follows from the homogeneity of F_1). Thus the paper's headline physical claim (massless all-coupling infinite-lattice gauge theory) is likely correct, while one of its advertised 'two constructions' is invalid as written. The reader's CONDITIONAL verdict is therefore appropriate; no verdict change.","tokens_in":14205,"tokens_out":30338,"duration_ms":310617,"concrete_test":"Perform a direct norm computation for one subdivision step in d=4. Take L=Z^4, h=1, L' the refinement with h'=1/2. Define β' on a large finite box by β'_y=2x on vertical edges, β'_x=0 on horizontal edges, and taper β' to zero near the boundary. Set α'=dβ' and compute ||α'||^2 = 4M and ||P^2_{L,L'} α'||^2 = 16M, where M is the number of coarse xy-plaquettes in the box. If the ratio is 4 as predicted, the 'contraction' inequality fails. (Alternative: compute the Fourier symbol of the subdivision map; its operator norm on ℓ^2 is >1.)","verdict_should_be":"UNCHANGED","load_bearing_attack":"Section 5.4 asserts that for a subdivision L' of L, the linear map (P^2_{L,L'})• : Im d^g_{L',1} → Im d^g_{L,1} (summing each coarse 2-cell's fine sub-cells) is a contraction w.r.t. the inner products h^{d-4} times the Euclidean one. This is false for every d>2. Counterexample: take d=4, h=1, L=Z^4, L' the half-mesh refinement. Let β' be a 1-cochain on L' with β'_y = 2x on vertical edges and β'_x = 0 on horizontal edges, truncated to finite support; set α' = dβ'. Then α' ≡ 1 on every fine square in the xy-plane, so on a region of M coarse xy-plaquettes ||α'||^2_{L'} = 4M. Its image α = P^2 α' has value 4 on each coarse plaquette, giving ||α||^2_L = 16M. Thus the squared operator norm is at least 4 (norm factor 2), not ≤1. The same calculation with h' = h/2 yields squared norm 4·2^{d-4} > 1 for all d>2. Hence (P^2_{L,L'})• is not a contraction, the system (5.10) is not a projective system in LG (see Definition 5.1, which requires Φ a contraction), and the limit measure μ_{β,∞} of §5.4 is not defined. This gap does not affect Theorem 5.7.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proposes two constructions of infinite-volume and continuum limits for Abelian polyhedral gauge theories with a modified Villain action. The first construction (Section 4) renormalizes the inner products on the spaces of exact Lie-algebra-valued 2-cochains so that the pullback maps become co-isometries, yielding a projective system of heat-kernel measures and hence a limit measure. The second construction (Section 5) works at the level of projective limits of Hilbert spaces in a category LG of compact Abelian groups, Hilbert spaces, and character maps; it produces a measure μ_{β,Z^d} on the infinite lattice Z^d for which the connected two-point function of Wilson loops is shown to decay no faster than n^{-d} for every β>0 and every d>2 (Theorem 5.7). The paper then attempts to extend the second construction to a Euclidean-invariant continuum limit over all infinite cubical lattices in R^d (Section 5.4). The central contrast claimed is that the standard thermodynamic limit of Abelian lattice gauge theory is massless only for small couplings, whereas the projective-limit measure is massless at all couplings.","tokens_in":14694,"tokens_out":30177,"duration_ms":263088,"significance":"If Theorem 5.7 is correct, the paper gives a conceptually clean and explicit construction of a translation-invariant, all-coupling massless phase for Abelian gauge theory in d>2, complementing the known small-coupling massless phase and large-coupling confinement picture. The Fourier computation in Proposition 5.5 is transparent and the use of projective limits of Hilbert spaces is elegant. Theorem 4.1, which renormalizes inner products to make subdivision maps co-isometries, is a sound and useful general mechanism. These strengths are substantial. However, two load-bearing points need repair: the contraction property asserted in Section 5.4 is factually false for the stated inner-products, and the proof of Lemma 5.6 contains a false smoothness claim. The main infinite-lattice masslessness theorem may survive, but the continuum-limit construction in Section 5.4 as written is not valid.","major_comments":[{"comment":"The assertion that the subdivision maps (P^2_{L,L'})• are contractions with respect to the h^{d-4}-scaled inner products is false for every d>2. Take L=Z^4 with mesh 1 and L' the half-mesh refinement. Let β' be the fine 1-cochain with β'_y=2x on vertical edges and β'_x=0 on horizontal edges. Then α'=dβ' equals 1 on every fine square in the xy-plane. On a region of M coarse xy-plaquettes, ||α'||²_{L'}=4M, while P^2α' equals 4 on each coarse plaquette, so ||P^2α'||²_L=16M. For general d the squared norm ratio is 4·2^{d-4}>1. Thus (P^2_{L,L'})• is not a contraction, (5.10) is not a projective system in LG (Definition 5.1 requires the Hilbert-space map to be a contraction), and the measure μ_{β,∞} of §5.4 is not constructed. This does not affect Theorem 5.7.","section":null},{"comment":"The proof states 'χF0 − F1 is smooth'. This is false. For example, when d=3 and I={i,j}, the leading singularity of F0−F1 at the origin is a homogeneous degree-2 function that is not a quadratic polynomial; along the direction (1,0,0) it vanishes to second order, while along (1,2,1) the second directional derivative is nonzero. Hence the function is not C² at 0. Since the proof uses smoothness to transfer the n^{-d} decay from F1 to χF0, the argument as written is incomplete. The lemma's conclusion may still be true by a more careful homogeneous-distribution analysis, but the present proof needs a corrected argument.","section":null},{"comment":"The paper identifies the projective limit of (Im d^g_{K,1}, (P^2_{K,K'})•) in Hilb1 with the image Im d^{Z^d}_1 of the coboundary on ℓ2 cochains and then uses an orthogonal projection Π onto this image. For d>2 the range of d_1: ℓ2(Λ^1)→ℓ2(Λ^2) is not closed: the symbol m(ξ)∧ vanishes at ξ=0, and a section in the pointwise image need not have a square-integrable preimage because |m(ξ)|^{-1} is not bounded. Thus 'Im d^{Z^d}_1' must be understood as the closure of the range, or Π is not a well-defined orthogonal projection onto a closed subspace. The formulas in Proposition 5.5 remain valid if Π is interpreted as projection onto the closure, but this needs to be stated and proved.","section":null}],"minor_comments":[{"comment":"The abstract's claim of 'two constructions of continuum and thermodynamic limits' is stronger than what is currently established: the Euclidean-invariant continuum limit in §5.4 is not constructed because of the contraction failure. The abstract and Section 5 should be revised to state the continuum-limit claim conditionally or to replace the inner products so that the contraction property holds.","section":null},{"comment":"The phrase 'not faster than n^{-d}' should be clarified; the intended meaning is a lower bound |...| ≥ c n^{-d} for all large n (or infinitely many n). This would remove a possible ambiguity about whether the authors claim a lower or an upper bound.","section":null},{"comment":"The formula in the displayed integral has a stray 'ξ M' where the matrix element is meant. This is a typographical issue but should be corrected.","section":null}],"recommendation":"major_revision","confidential_remarks":"The contraction counterexample in §5.4 is decisive and should be communicated clearly: the infinite-volume Euclidean-invariant continuum limit, as constructed, is not a projective system in the category LG. The Lemma 5.6 proof also has a substantive gap. I would advise the editor to require the authors either to repair these two points or to explicitly withdraw the affected claims. The main infinite-lattice masslessness theorem may still be salvageable, so rejection does not seem warranted at this stage."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Here's my take. The headline result, Theorem 5.7, is good: it constructs a translation-invariant measure on the infinite cubical lattice whose Wilson-loop two-point function decays no faster than n^{-d} for all β>0 and d>2. The construction via projective limits of Hilbert spaces is new, at least for d>2, and the proof is a transparent Fourier calculation. The d=2 case reduces to prior work, and the self-citations to Part I are legitimate.\n\nSection 5.4, however, is not just missing a proof — the contraction assertion is false. For d=4, L=Z^4, L' the half-mesh refinement, take β'_y=2x on vertical 1-cochains and β'_x=0. Then α'=dβ' is 1 on every fine xy-square, so ||α'||^2=4M over M coarse squares. The subdivision map P^2 sums the fine values, giving α=4 on each coarse square, so ||α||^2=16M. With the h^{d-4} scaling the squared norm ratio is 4·2^{d-4} for d>2, so (P^2_{L,L'})• is not a contraction, (5.10) is not a projective system in LG, and μ_{β,∞} is not constructed. Theorem 5.7 does not depend on this, so the central infinite-lattice result stands. There are also minor notational slips, e.g. in (3.2).\n\nThis paper deserves a serious referee. The referee should focus on §5.4; if the contraction cannot be repaired, the Euclidean-invariant continuum limit should be removed or substantially revised. The infinite-lattice theorem in §5.2–5.3 is worth engaging with, and the counterexample to the contraction claim is instructive in its own right.","headline":"The infinite-lattice masslessness theorem is likely right, but the Euclidean-invariant continuum limit in §5.4 rests on a false contraction claim.","tokens_in":15093,"tokens_out":6561,"would_cite":true,"duration_ms":55830,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["81T13","81T08","28C20"],"pacs":[],"model":"deepseek-v4-flash","headline":"A projective-limit construction yields a translation-invariant Abelian lattice gauge theory that remains massless at every coupling in dimensions above two.","keywords":["Abelian gauge theory","Villain action","projective limit","heat kernel measure","infinite lattice limit","massless phase","Wilson loop","constructive quantum field theory"],"falsifier":"Compute the norm of the subdivision map (P^2_{L,L'})• on a single exact 2-cochain on a subdivided cube in d=4 using the h^{0}-scaled Euclidean inner product; if the norm of the image exceeds the norm of the original cochain, the contraction property fails and the continuum-limit projective system is invalid.","tokens_in":14117,"feed_emoji":"⚛️","tokens_out":5891,"duration_ms":50451,"temperature":0.7,"pith_summary":"This paper aims to show that Abelian polyhedral gauge theories, such as lattice U(1) gauge theory, admit continuum and infinite-volume limits built from projective systems of heat-kernel measures. The central result is that on the infinite cubical lattice in dimensions d>2, the resulting translation-invariant measure describes a massless gauge field for every value of the coupling β, unlike the standard thermodynamic limit, which is massless only for small couplings. A second, Euclidean-invariant continuum limit is also constructed by iterated subdivision, subject to a contraction property that is asserted rather than proved. If correct, the construction gives a well-defined all-coupling massless phase and a new route from lattice to continuum gauge theories.","feed_headline":"New lattice gauge construction yields massless phase at every coupling","feed_subtitle":"A new projective-limit construction gives a translation-invariant, all-coupling massless phase in d>2.","key_machinery":"The key object is the modified Villain measure, defined by pulling back the heat kernel on the closed subgroup Im d_1 of 2-cochains along the coboundary, then pushing forward by the inverse of the induced map on gauge-equivalence classes. The carrying machinery is the category LG of triples (compact Abelian group, Hilbert space, and a homomorphism from the dual group into the Hilbert dual) whose Fourier-analytic heat kernels exp(-4π²β||J(χ)||²) allow projective limits of heat-kernel measures; together with the identification of the infinite-lattice limit with the image of the ℓ² coboundary operator, this reduces correlation calculations to Fourier multipliers on the torus and the orthogonal","core_discovery":"For a finite contractible polyhedral complex, the modified Villain action—the heat kernel composed with the coboundary operator—pushes forward to a heat kernel measure on the image of the coboundary, and inverting the coboundary on gauge equivalence classes identifies the gauge theory measure with that heat kernel measure. Renormalizing inner products on the image spaces makes refinement and restriction maps co-isometries, yielding a projective system of measures (Theorem 4.1) and hence a limit measure. On the infinite lattice Z^d, the measure is lattice-translation-invariant and the connected two-point function of Wilson loops decays only as a power n^{-d} (Theorem 5.7), the signature of a","pith_inferences":["If the massless-all-coupling property survives in the continuum limit, this projective-limit gauge theory may inhabit a different phase from the standard Wilson/Villain thermodynamic limit, possibly one without a confining phase.","The Fourier-multiplier technique used to prove polynomial decay of Wilson-loop correlations could be extended to higher-point functions to test whether the massless phase exhibits Gaussian behavior or interacting corrections.","The unproved contraction property could be tested numerically on simple subdivisions; if it fails only for d≥4, the Euclidean-invariant continuum construction might still hold for d=3, the physically most relevant case.","The framework may extend to non-Abelian gauge groups by replacing heat kernels on image subgroups with heat kernels on more general homogeneous spaces, a direction the paper leaves to a sequel."],"forward_implications":["The infinite-lattice measure μ_{β,Z^d} gives a translation-invariant Abelian lattice gauge theory whose Wilson-loop correlations decay polynomially, not exponentially, for every coupling in d>2, so the construction exhibits a massless phase with no coupling restriction.","The projective-system theorem provides a general existence result for continuum and infinite-volume limits of Abelian polyhedral gauge theories in arbitrary dimension, extending the d=2 consistent-measure constructions to d>2.","For d=2 the construction reduces to the usual thermodynamic limit of U(1) lattice gauge theory, with O_β(p,q)=0 unless p=q, matching known results.","The Euclidean-invariant continuum limit measure μ_{β,∞}, if the contraction step is valid, is invariant under the full Euclidean group E(d), so the limiting field theory is rotation- and translation-symmetric."],"fun_headline_variants":["Projective limits make Abelian gauge theory massless at all couplings","All-coupling massless phase in Abelian lattice gauge theory","Heat-kernel projective system yields massless gauge phase","Abelian lattice gauge: massless for every coupling value"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The Euclidean-invariant continuum limit rests on the unproved assertion that subdivision maps are contractions under dimension-scaled inner products; if that assertion fails, the continuum-limit measure is not constructed, though the infinite-lattice masslessness result stands independently.","fun_headline_variants_meta":{"raw":{"variants":["Projective limits make Abelian gauge theory massless at all couplings","All-coupling massless phase in Abelian lattice gauge theory","Heat-kernel projective system yields massless gauge phase","Abelian lattice gauge: massless for every coupling value"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000156,"raw_usage":{"total_tokens":983,"prompt_tokens":603,"completion_tokens":380,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":347,"completion_tokens_details":{"reasoning_tokens":323}},"tokens_in":347,"tokens_out":380,"duration_ms":3781,"temperature":1.0,"reasoning_tokens":323,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-04T08:17:04.667286+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the norm of the subdivision map (P^2_{L,L'})• on a single exact 2-cochain on a subdivided cube in d=4 using the h^{0}-scaled Euclidean inner product; if the norm of the image exceeds the norm of the original cochain, the contraction property fails and the continuum-limit projective system is invalid.","supporting_citations":[],"review_version":1}