{"id":"03a83f55-e5a9-4bff-bf52-af59beb361df","arxiv_id":"2506.17813","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Near flat singularities, two-dimensional area-minimizing mod(q) currents in codimension one are C^{1,alpha}-perturbations of graphs of radially homogeneous harmonic multiple-valued functions, and top-density flat singularities are isolated in all codimensions.","lead":"This paper describes the fine local shape of singular points on area-minimizing surfaces counted modulo q, a model for soap films. It proves that flat singular points in three dimensions are small smooth distortions of a wave pattern built from sine functions in alternating sectors, and that the top-density flat singularities are isolated.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Assumption 4.3 rests on Corollary 4.2, whose proof is omitted; if the cited [3, Props 2.3–2.4] do not transfer to codimension >1, the variational and competitor arguments of Sections 5–6 have no global foundation.","rationale":"I agree with the reader's identification of the weakest assumption. The body of the paper is organized so that every estimate in Sections 5 and 6 is stated under Assumption 4.3: Proposition 5.1, Corollary 5.2, Proposition 5.4, Lemma 6.5, Lemma 6.4, and Theorem 6.2 all begin with the same conditional hypothesis. Thus the entire frequency-decay machine is contingent on Corollary 4.2. The paper delegates this structurally nontrivial generalization to an omitted 'same reasoning' argument, so the concern is not manufactured; it is the unique place where a load-bearing step is asserted rather than proved. I checked Theorem 4.1 for internal coherence: the eigenvalue argument is terse, but the connectedness of each Ω^± component plausibly eliminates interior zeros and forces the sector length π/α, so I do not see an outright inconsistency there. I also do not find a more serious gap in the variational or competitor arguments of Sections 5–6, which are written out in substantial detail. The decisive unknown is whether the codimension-one proof of [3, Propositions 2.3–2.4] really transfers to n̄ > 1 with only the two-dimensional classification and the weaker tilt excess decay. Since this is exactly the reader's concern, I agree with the CONDITIONAL verdict and do not recommend changing it.","tokens_in":23898,"tokens_out":15913,"duration_ms":174483,"concrete_test":"Independently reconstruct the proof of Corollary 4.2 for n̄ = 2, Q = 2: starting from Theorem 4.1 and [8, Proposition 7.2], carry out the argument of [3, Propositions 2.3–2.4] step by step, writing down the induction over intervals of flattening and the Whitney decomposition, and record every occurrence of the codimension-one hypothesis n̄ = 1. Then verify the claimed containment F_Q(T) ∩ B_η ⊂ Φ(Γ) ⊂ M and the estimate Eno(T, B_r) ≤ C m_0 r^{2-2δ} hold in that case. If the proof cannot be completed without an additional assumption on the normal approximation, Assumption 4.3 is unjustified.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The reduction to Assumption 4.3 is the hinge of the paper: once a single center manifold M and M-normal approximation N are available on all of B_{3/2}, Proposition 5.4 and Lemma 6.5 can be applied, and Theorem 6.2 yields the power-law decay of H, D, and I from which Theorems 1.2 and 1.3 are drawn. That reduction is exactly Corollary 4.2, whose proof the paper explicitly omits: \"We omit the proof of Corollary 4.2...\" in Section 4. The stated justification is that [3, Propositions 2.3 and 2.4] are codimension-independent once Theorem 4.1 and a weaker power-law tilt excess decay [8, Proposition 7.2] replace [4, Theorem 1.3]. This is an assertion about a nontrivial induction over intervals of flattening, Whitney regions, and the absence of lower-density flat singularities accumulating near the origin; it is not demonstrated. If the cited propositions do not transfer, then m_{0,j} may not be controlled on a single interval, the estimate Eno(T, B_r) ≤ C m_0 r^{2-2δ} can fail before Φ(Γ) is reached, and the global competitor construction in Section 6 has no known domain of admissibility. The main theorems would then not follow from the given arguments.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript proves two structural results for two-dimensional area-minimizing currents modulo an even integer q. Under Assumption 1.1 with codimension n-bar=1, Theorem 1.2 asserts that near a flat singularity of density Q, the current is a C^{1,alpha}-perturbation of the multigraph of an explicit radially homogeneous special multiple-valued function u built from harmonic polynomials r^{I0} sin(I0 theta) on alternating sectors. Theorem 1.3 asserts in any codimension that the set F_Q(T) of flat singularities of density Q is discrete. The proof combines the center manifold and M-normal approximation machinery of [5], a new classification of two-dimensional homogeneous tangent functions (Theorem 4.1), variational estimates and almost Dir-minimality for the reparameterized normal approximation (Section 5), and a competitor construction that yields power-law frequency decay (Theorem 6.2).","tokens_in":24215,"tokens_out":5037,"duration_ms":43425,"significance":"If valid, these results give the first fine description of the local structure of two-dimensional mod(q) area-minimizing currents at top-density branch points in codimension one, and they establish discreteness of such points in arbitrary codimension. The paper is careful in attributing prior machinery and the strategy is coherent; in particular, the reduction to explicit homogeneous harmonics and the clean statement of Theorem 1.3 are significant advances. The main shortcomings are not conceptual but expository/completeness: several load-bearing corollaries have omitted proofs, and the most delicate point in Theorem 4.1 is compressed.","major_comments":[{"comment":"The reduction to a single center manifold and M-normal approximation on B_{3/2}, formalized in Assumption 4.3, is exactly the content of Corollary 4.2, whose proof is omitted. The statement that 'no part of the proofs therein rely on the codimension being 1' is an assertion about a multi-step induction over intervals of flattening, Whitney regions, and control of m_{0,j}; it is not demonstrated. Since Proposition 5.4, Lemma 6.5, and Theorem 6.2 are all stated under Assumption 4.3, the central frequency-decay argument in Section 6 has no proven global foundation if Corollary 4.2 fails. The authors should either include the proof of Corollary 4.2 or provide a precise verification that [3, Propositions 2.3 and 2.4] transfer to codimension n-bar > 1 under the hypotheses of Theorem 4.1 and [8, Proposition 7.2].","section":"Section 4, Corollary 4.2 and following paragraph"},{"comment":"Corollary 5.2, the almost-monotonicity estimate I'(r) >= -C r^{gamma-1}, is used in the proof of Theorem 6.2 to select r_1, to control I(r)-alpha from below, and to integrate the differential inequality; it is therefore load-bearing. Its proof is omitted, with only a reference to [12, Proof of Theorem 3.2] and [9, Lemma 4.1]. Because the definition of I(r) in this paper is the classical (non-regularized) frequency for the reparameterized map N, the transfer from the regularized frequency in [12, 5] is not automatic; at least a sketch of the computation of I'(r) using (5.2)-(5.4) should be supplied.","section":"Section 5, Corollary 5.2"},{"comment":"The proof that p is harmonic across the nodal set Omega^0 is summarized in a few sentences. The key step asserts that a tangent function g at x is translation-invariant in the direction spanned by x, reduces to a one-dimensional homogeneous Dir-minimizer h, and that the inner variation identity implies |Dh| constant, from which the equality of normal derivatives partial_nu p_+(x) = partial_nu p_-(x) is deduced. Several implications require justification: why the tangent function at a point x != 0 with u(x) = QJ0K is translation-invariant along x; why constancy of |Dh| implies equality (not just equality of absolute values) of the normal derivatives of p_+ and p_-; and how the argument handles points where more than two nodal lines meet. Since Theorem 4.1 is used both in Corollary 4.2 and in the classification underlying Theorem 6.2, this step needs to be written out in full.","section":"Section 4, Theorem 4.1, harmonicity of p"}],"minor_comments":[{"comment":"The notation N is used both for the map on M and for its reparameterization N := N circ Phi; the distinction is sometimes blurred in the displayed formulas (e.g., Proposition 5.1 uses |DN|^2 on Phi(B_r) and on B_r). Please use different symbols or state explicitly which domain is meant in each integral.","section":"Section 5, first paragraph"},{"comment":"The line 'by Theorem [7, Theorem 11.5]' contains misplaced brackets; also the proof of (6.10) is only sketched via contradiction and would benefit from more details on how the eigenfunction expansions pass to the limit.","section":"Section 6, proof of Lemma 6.5"},{"comment":"The statement 'there exist i != j such that v_i(theta) != v_j(theta) for each theta in I' should be 'for almost every theta in I' unless the selections are continuous; otherwise it is not justified from W^{1,2} selections alone.","section":"Section 4, proof of Theorem 4.1"},{"comment":"There is a typo: 'for abritrary r' should be 'for arbitrary r'.","section":"Section 6, proof of Theorem 6.2"},{"comment":"The sentence 'This allows us to assume that we are working with a single center manifold' refers to Corollary 4.2, but the reader is not told until Section 4 that the proof of that corollary is omitted; consider moving or expanding the discussion to flag this dependency.","section":"Section 2, Strategy of proof"}],"recommendation":"major_revision","confidential_remarks":"The paper's main theorems are plausible and the architecture is standard, but the two omitted corollaries and the compressed harmonicity argument in Theorem 4.1 are exactly the places where the argument could fail. I would be inclined to accept after the authors provide the missing proofs or at minimum a detailed verification of the transfer of [3, Propositions 2.3-2.4] and [12, Theorem 3.2]. Also, the companion paper [17] is cited as 'forthcoming' for a step in Remark 1.4; please ensure the results do not depend essentially on an unpublished reference."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: this is a real step forward in the mod(q) regularity program, and I think the main theorems are probably true. The catch is that the proof of Corollary 4.2—the reduction to a single center manifold, which everything downstream depends on—is omitted, and that omission is load-bearing, not cosmetic.\n\nWhat is genuinely new: Theorem 1.2 gives the first C^{1,α} harmonic multigraph structure near flat singularities for 2D mod(q) currents in codimension one, and Theorem 1.3 gives discreteness of top-density flat singularities in arbitrary codimension. Theorem 4.1, the classification of 2D tangent functions, is also new beyond the codimension-one setting. The overall architecture is sensible: classify tangent functions, force a single center manifold, derive variational estimates and almost Dir-minimality, construct a competitor, and get frequency decay. The estimates in Sections 5 and 6 are detailed and the reliance on prior work is explicit rather than hidden.\n\nThe main soft spot is exactly what the stress-test flags. Corollary 4.2 is used to pass from the local tangent classification to the global Assumption 4.3, and Sections 5–6 operate entirely under that assumption. The paper says the proof is omitted because the arguments in [3, Props 2.3–2.4] do not use codimension one once Theorem 4.1 and the power-law tilt excess decay from [8, Prop 7.2] are available. That is plausible, but it's a nontrivial transfer: the interval of flattening, Whitney regions, and the control of m_{0,j} on a single interval are exactly what need checking. A referee should ask for this proof to be written out, or at least for a detailed outline that verifies the higher-codimension steps. Corollary 5.2 also has no proof, but there it's more believable that it follows routinely from Proposition 5.1 plus the cited references. The harmonicity step in Theorem 4.1 is compressed—the tangent-function argument across the nodal set is stated rather than fully shown—but I don't see an obvious error.\n\nWho is this for: GMT specialists working on mod(q) regularity, singular sets, and multiple-valued functions. They will want to read it carefully. It deserves peer review, not desk rejection. My recommendation: send it to a serious referee, and make it a condition of acceptance that the proof of Corollary 4.2 be included in full or the paper be restructured so the main theorems do not depend on an unproved assertion.","headline":"Likely correct and genuinely new structural result for 2D mod(q) minimizers, but the proof of the key reduction to a single center manifold is omitted and must be supplied.","tokens_in":24738,"tokens_out":2530,"would_cite":true,"duration_ms":26575,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["49Q20","49Q15"],"pacs":[],"model":"deepseek-v4-flash","headline":"Near flat singularities, two-dimensional mod(q) area minimizers are C^{1,α} perturbations of explicit alternating harmonic multigraphs.","keywords":["mod(q) area-minimizing currents","flat singularities","branch points","Q-valued functions","frequency function","harmonic polynomials","codimension one regularity","singular set discreteness"],"falsifier":"As a direct test, look at the rescaled normal approximation $N_r = (N\\circ\\Phi)(r\\,\\cdot)/D(r)^{1/2}$ on $\\partial B_1$ near a flat singularity: Theorem 6.2 predicts the angular spectrum is dominated by one integer mode $I_0 \\ge 2$, with all other modes decaying as $r^\\gamma$. A current whose leading angular mode has non-integer homogeneity, or a flat singularity whose blow-up is not made of alternating sectors $r^{I_0}\\sin(I_0\\theta)$, would contradict the classification in Theorem 4.1 and collapse Theorem 1.2.","tokens_in":23723,"feed_emoji":"📐","tokens_out":10817,"duration_ms":102255,"temperature":0.7,"pith_summary":"The paper establishes a sharp local model for two-dimensional area-minimizing surfaces modulo an even integer $q$, at points where the surface has a flat singularity: a tangent cone that is a single plane of multiplicity $Q = q/2$. In codimension one, the main theorem says that, up to rotation, the surface in a small ball is a $C^{1,\\alpha}$-perturbation of the multigraph of an explicit function built from alternating sectors of the harmonic polynomial $r^{I_0}\\sin(I_0\\theta)$, with integer frequency $I_0 \\ge 2$ and real coefficients. The same analysis also gives, in every codimension, uniqueness of the fine blow-up with a power-law rate of decay, and shows that the density-$Q$ flat singularities that are genuinely mod$(q)$ form a discrete set. A reader should care because this is the first complete local description of branch-point singularities in this class of variational surfaces.","feed_headline":"Mod-q minimizers near flat points are harmonic fans","feed_subtitle":"The paper pins down the local geometry of branch points in two-dimensional mod-q area minimizers.","key_machinery":"The load-bearing object is the pair consisting of a center manifold $M$, a two-dimensional surface near which the current is concentrated, and its associated $M$-normal approximation $N$, a special $Q$-valued map that records the sheets of the current over $M$. On this parametrization the paper studies the frequency function $I(r) = rD(r)/H(r)$, where $D(r)$ is the gradient energy of $N$ on the ball of radius $r$ and $H(r)$ is the $L^2$ height on the boundary circle; a frequency that is almost constant forces the map to be almost homogeneous. The main technical step is a comparison with a specially chosen competitor built from eigenfunctions of the Laplace operator on the boundary arcs, which yields the almost-minimality of $N$ and hence the key decay estimate of Theorem 6.2 for $I(r)$, $H(r)/r^{2I_0+1}$, and $D(r)/r^{2I_0}$. The whole argument requires a single such center manifold covering the ball, which is encoded in Assumption 4.3.","core_discovery":"The central claim is Theorem 1.2: under the paper's standing assumptions, with $q = 2Q \\ge 4$ and codimension $\\bar n = 1$, there are $r_0 > 0$ and $\\alpha \\in (0,1)$ such that, up to rotation, $T$ restricted to $B_{r_0}$ is a $C^{1,\\alpha}$-perturbation of the multigraph whose height is $u(r,\\theta) = \\sum_{i=1}^Q J c^{\\pm}_{j,i} r^{I_0}\\sin(I_0\\theta)K$ on the alternating sectors $U_j^\\pm$, where $I_0 \\in \\mathbb N$, $I_0 \\ge 2$, and the coefficients are real. The multigraph is special in that each sheet carries a sign $+1$ or $-1$ according to the sector, reflecting the mod$(q)$ structure. Theorem 1.3 states that, in any codimension, the set $F_Q(T)$ of flat singularities of density $Q$ where the current is genuinely mod$(q)$ is discrete. The proofs rest on a frequency-function decay estimate that yields power-law convergence of rescalings to the harmonic model.","pith_inferences":["Beyond the paper: if the same decay estimate holds in higher codimension once lower-density branch points are ruled out, the local model at any flat singularity would be the same alternating harmonic fan, regardless of codimension.","The classification of tangent functions as $r^{I_0}\\sin(I_0\\theta)$ suggests a direct connection to nodal-set geometry of harmonic polynomials; the singular set near a flat singularity should be a union of smooth arcs with angles $\\pi/I_0$, a structure that might be detected numerically in soap-film models.","A quantitative strengthening the authors do not attempt: extract explicit bounds for the exponent $\\alpha$ and the decay rate $\\gamma$ in terms of $q$ and the ambient dimension; the proof only establishes existence."],"forward_implications":["Every density-$Q$ flat singularity in a codimension-one two-dimensional mod$(q)$ minimizer has a unique blow-up that is an explicit alternating harmonic fan with integer homogeneity $I_0 \\ge 2$.","The rescaled currents converge to that fan with a power-law rate, not merely qualitatively, giving quantitative control on the local shape.","In any codimension, density-$Q$ flat singularities that are genuinely mod$(q)$ cannot accumulate; each one is isolated.","Near such a singularity, the singular set has a rigid geometry: the model's nodal lines $\\theta = k\\pi/I_0$ imply the surface is smooth away from a finite union of arcs meeting at the singularity."],"supporting_citations":[{"why":"Constructs the center manifold and the M-normal approximation for mod(q) minimizers, on which the paper's frequency analysis takes place.","marker":"[5]"},{"why":"Gives the codimension-one tangent-function classification and the propositions that Corollary 4.2 invokes to reduce to a single center manifold.","marker":"[3]"},{"why":"Supplies the fine blow-up and uniqueness theory for mod(q) singularities used in Theorem 1.3 and in the compactness argument.","marker":"[7]"},{"why":"Provides the selection and eigenvalue structure for Q-valued maps used in the classification of homogeneous minimizers and in the blow-up argument.","marker":"[10]"},{"why":"Is the source of the variational estimates for the frequency function and of the competitor strategy adapted here.","marker":"[12]"},{"why":"Gives the almost Dir-minimality and decay framework for two-dimensional almost-minimal currents that the paper adapts.","marker":"[15]"},{"why":"Provides the C^{1,α} perturbation theorem for lower-density points used in the proof of Theorem 1.2.","marker":"[2]"},{"why":"Supplies the excess decay for mod(2Q) hypercurrents that Corollary 4.2 needs in higher codimension.","marker":"[4]"}],"fun_headline_variants":["Harmonic fans describe mod-q branch points, isolated","Near flat points, mod-q minimizers are harmonic fans","Mod-q area minimizers: flat singularities are harmonic and isolated","Codim-one mod-q minimizers: harmonic structure near flat singularities","Mod-q branch points: harmonic multigraphs, isolated"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The entire local model rests on Assumption 4.3, namely that after the preliminary analysis the current is captured on a single center manifold with one normal approximation over the whole unit ball; the paper obtains this from Corollary 4.2, whose proof is omitted and which is asserted to follow from [3, Propositions 2.3 and 2.4].","fun_headline_variants_meta":{"raw":{"variants":["Harmonic fans describe mod-q branch points, isolated","Near flat points, mod-q minimizers are harmonic fans","Mod-q area minimizers: flat singularities are harmonic and isolated","Codim-one mod-q minimizers: harmonic structure near flat singularities","Mod-q branch points: harmonic multigraphs, isolated"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001415,"raw_usage":{"total_tokens":5699,"prompt_tokens":915,"completion_tokens":4784,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":531,"completion_tokens_details":{"reasoning_tokens":4699}},"tokens_in":531,"tokens_out":4784,"duration_ms":35340,"temperature":1.0,"reasoning_tokens":4699,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T19:00:30.922364+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"As a direct test, look at the rescaled normal approximation $N_r = (N\\circ\\Phi)(r\\,\\cdot)/D(r)^{1/2}$ on $\\partial B_1$ near a flat singularity: Theorem 6.2 predicts the angular spectrum is dominated by one integer mode $I_0 \\ge 2$, with all other modes decaying as $r^\\gamma$. A current whose leading angular mode has non-integer homogeneity, or a flat singularity whose blow-up is not made of alternating sectors $r^{I_0}\\sin(I_0\\theta)$, would contradict the classification in Theorem 4.1 and collapse Theorem 1.2.","supporting_citations":[{"cited_title":"De Lellis, J","cited_arxiv_id":null,"evidence_quote":"Constructs the center manifold and the M-normal approximation for mod(q) minimizers, on which the paper's frequency analysis takes place."},{"cited_title":"De Lellis and E","cited_arxiv_id":null,"evidence_quote":"Provides the selection and eigenvalue structure for Q-valued maps used in the classification of homogeneous minimizers and in the blow-up argument."},{"cited_title":"of Math.(2) (2016), 577–617","cited_arxiv_id":null,"evidence_quote":"Is the source of the variational estimates for the frequency function and of the competitor strategy adapted here."},{"cited_title":"Differential Geom","cited_arxiv_id":null,"evidence_quote":"Gives the almost Dir-minimality and decay framework for two-dimensional almost-minimal currents that the paper adapts."},{"cited_title":"247 (2024), Paper No","cited_arxiv_id":null,"evidence_quote":"Supplies the excess decay for mod(2Q) hypercurrents that Corollary 4.2 needs in higher codimension."}],"review_version":1}