{"id":"1930eba4-d404-46f7-8a7e-7095e8824270","arxiv_id":"2508.05230","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"For Holder exponents greater than 1/2, weak C^{1,theta} solutions of the Darboux equation are equivalent to C^{1,theta} isometric immersions, via a new distributional Gaussian curvature and flatness criterion.","lead":"This paper defines a new notion of weak solutions for the Darboux equation, a PDE tied to how 2D surfaces bend into 3D space, valid when the surface has only a mild amount of smoothness (Holder exponent above 1/2). It proves that these weak solutions are exactly equivalent to low-regularity isometric immersions, extending a classical correspondence and providing a flatness criterion for rough metrics.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Lemma 4.1's stability proof in Appendix B appears to use an incorrect Beltrami difference equation; the C^{1,θ} chart continuity (4.1) is not established as written.","rationale":"The paper's central theorem is plausible and internally consistent: the weak Darboux formulation, the distributional curvature, and the flatness criterion interact naturally, and the factor-2 discrepancy in Def. 4.2 is inconsequential for the zero/nonzero conclusions. The reader's weakest assumption correctly identified Lemma 4.1 as the pivotal technical input. My stress-test pinpoints a specific flaw in the proof of that lemma: the displayed difference equation in Appendix B has an incorrect source term, making the claimed C^{1,θ} stability unproved. This does not warrant rejection because the result is very likely true and the proof repairable, but it does reinforce the CONDITIONAL verdict: the authors must correct Lemma 4.1's proof (or replace it with an explicit citation for the stability of the principal Beltrami solution) before the main theorem is fully rigorous. I therefore leave the reader's verdict unchanged.","tokens_in":14636,"tokens_out":30510,"duration_ms":324795,"concrete_test":"Re-derive the equation for f_ε = ψ_ε − ψ by subtracting the two Beltrami equations. If the correct source term is (μ_ε − μ)ψ_z (or (μ_ε − μ)ψ_{ε,z}), verify that Theorem 15.0.6 of [2] yields [Df_ε]_{C^{0,θ}} ≤ C ||μ_ε − μ||_{C^{0,θ}}, giving (4.1). If the printed equation with source (μ − μ_ε)f_ε_z is used instead, show that the Schauder estimate is inapplicable because the source is not an independent inhomogeneity. This analytical check settles whether Lemma 4.1's stability estimate is proved.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The proof of Theorem 1.1 depends on the distributional curvature of Prop. 4.2, whose continuity in the metric relies on Lemma 4.1, specifically the stability estimate (4.1). In Appendix B, the difference f_ε = ψ_ε − ψ is claimed to satisfy f̄ε_z = μ f_ε_z + (μ − μ_ε) f_ε_z. Subtracting the Beltrami equations for ψ and ψ_ε gives instead f̄ε_z = μ_ε f_ε_z + (μ_ε − μ) ψ_z (equivalently μ f_ε_z + (μ_ε − μ) ψ_{ε,z}). With the printed equation, the 'source' term (μ − μ_ε) f_ε_z depends on f_ε itself; it cannot be treated as an inhomogeneity in Theorem 15.0.6 of [2], and estimating its C^{0,θ} norm by ||μ − μ_ε||_{C^{0,θ}}||f_ε||_{C^{1,θ}} is circular and does not imply [Df_ε]_{C^{0,θ}} → 0. Without (4.1), the continuity assertion in Prop. 4.2 is unsupported, and the approximation arguments in both directions of Theorem 1.1 lack justification. The lemma is plausible and likely fixable, but as written the proof has a concrete gap. The separate factor-2 issue in Def. 4.2 does not affect the zero/nonzero distinction and is not the main obstruction.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper introduces a weak formulation of the Darboux equation for functions u ∈ C^{1,θ}(Ω), θ > 1/2, by interpreting the quotient det(∇²_gu)(1-|du|²_g)^{-3/2} dV_g as a distribution through a product of Hölder continuous functions. The main result, Theorem 1.1, asserts that u is such a weak solution if and only if u is the third component of a C^{1,θ} isometric immersion (ϕ,u): Ω → R³. The proof defines a distributional Gaussian curvature for C^{0,θ} metrics via conformal coordinates, proves a flatness criterion (Prop. 4.3), and then passes from smooth approximations u_ε to u using mollification and the continuity of the distributional curvature (Prop. 4.2).","tokens_in":14952,"tokens_out":13964,"duration_ms":160519,"significance":"If the proof is completed, this is a natural and valuable low-regularity extension of the classical Darboux correspondence, matching the threshold θ > 1/2 that already appears in weak formulations of the Gauss equation and in distributional product theory. The paper is also well structured: the distributional product Appendix A is essentially self-contained, the conformal chart machinery is standard, and both directions of the equivalence are addressed. The main theorem is a genuine contribution to the rigidity/flexibility discussion for low-regularity isometric immersions. However, the proof as written contains a concrete gap in the stability statement for conformal charts and a factor-of-two inconsistency in the definition of distributional Gaussian curvature; neither appears fatal, but both must be repaired before the paper is publishable.","major_comments":[{"comment":"The difference equation for f_ε = ψ_ε - ψ is incorrect. Subtracting ψ̄_z = μψ_z and ψ̄_{ε,z} = μ_εψ_{ε,z} gives f̄_{ε,z} = μ_ε f_{ε,z} + (μ_ε - μ)ψ_z, equivalently μ f_{ε,z} + (μ_ε - μ)ψ_{ε,z}. The printed equation f̄_{ε,z} = μ f_{ε,z} + (μ - μ_ε)f_{ε,z} has a source term that depends on f_ε itself, so Theorem 15.0.6 of [2] cannot be applied as an inhomogeneous estimate. The subsequent bound [(μ-μ_ε)f_{ε,z}] ≤ C‖μ-μ_ε‖‖f_ε‖ is circular and does not imply [Df_ε]_{C^{0,θ}} → 0. Moreover, local uniform convergence of f_ε together with a uniform C^{1,θ} bound does not by itself give convergence in C^{1,θ}. Since (4.1) is used in Prop. 4.2 and then in both directions of Theorem 1.1, this is a load-bearing gap. The correct equation should make it possible to apply Schauder estimates with a small source ‖μ_ε-μ‖‖ψ_z‖, but the proof must be rewritten.","section":"Appendix B, displayed equation for f_ε"}],"minor_comments":[],"recommendation":"major_revision","confidential_remarks":"The central theorem is plausible and the flawed step in Appendix B appears fixable by using the correct Beltrami difference equation and standard Schauder estimates. The factor-of-two issue in Definition 4.2 is also easily corrected. If the authors supply a correct proof of Lemma 4.1 and align the factor conventions, the paper could become acceptable. The repeated references and small typos are secondary. I recommend major revision rather than rejection because the main idea and the two-direction strategy are sound and the defects are local, not conceptual."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"This paper delivers something real: a weak formulation of the Darboux equation for u in C^{1,theta}, theta>1/2, and proves it is equivalent to existence of a C^{1,theta} isometric immersion. The distributional Gaussian curvature for C^{0,theta} metrics and the flatness criterion (Prop 4.3) are natural and useful. The overall strategy - view h = g - du^2 as a metric, show it is distributionally flat, then construct the immersion - is clean, and the main theorem is an honest new result. What is done well: the distributional product appendix is self-contained and essentially correct; Proposition 3.1 is well-posed; the flatness criterion is neat. The citation pattern is fine - self-citations appear only as context, and the main ingredients (Beltrami theory, Han-Hong) are independent. Now the soft spots. The factor-2 concern raised by the reader's report does not survive contact with the paper: from K_h = -2e^{-2sigma} Delta sigma and dV_h = e^{2sigma} dz, Definition 4.2 is exactly 2 times the integral of grad phi dot grad sigma, which is the classical Gaussian curvature. So no factor-2 error there. The real problem is in Appendix B, in the proof of Lemma 4.1. The difference f_epsilon = psi_epsilon - psi does not satisfy the displayed Beltrami equation. Subtracting psi_bar_z = mu psi_z from psi_epsilon_bar_z = mu_epsilon psi_epsilon_z gives f_bar_z = mu_epsilon f_z + (mu_epsilon - mu) psi_z, not mu f_z + (mu - mu_epsilon) f_z. As printed, the nonhomogeneous term depends on f_epsilon itself, so the appeal to Theorem 15.0.6 in [2] is circular and the estimate (4.1) does not follow. This matters: the continuity of the distributional curvature (Prop 4.2) and both directions of Theorem 1.1 rely on Lemma 4.1. I expect the lemma is true and the proof is fixable - standard Ahlfors-Bers stability should give it - but the current manuscript has not established it. So: the central idea is sound and the result is likely correct, but the written proof has a load-bearing gap in a key technical lemma. That is exactly what peer review is for. I recommend sending it to a serious referee, ideally one comfortable with Beltrami equations, and telling the authors to fix the difference equation and re-verify the stability estimate. After that, the paper should be publishable.","headline":"Genuinely new weak Darboux correspondence with a clean strategy, but the proof of Lemma 4.1 has a concrete Beltrami equation gap that must be fixed.","tokens_in":748,"tokens_out":2183,"would_cite":false,"duration_ms":59360,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["35D30","35J60","53C42"],"pacs":[],"model":"deepseek-v4-flash","headline":"The paper proves that, for $\\theta>1/2$, a $C^{1,\\theta}$ function solves the Darboux equation in a distributional sense if and only if it is the height function of a $C^{1,\\theta}$ isometric immersion of a two-dimensional metric into $\\mat","keywords":["Darboux equation","isometric immersions","low regularity","weak solutions","distributional Gaussian curvature","flatness criterion","conformal coordinates","Holder regularity"],"falsifier":"Construct a $C^{1,\\theta}$ function $u$ with $|du|_g<1$ that satisfies the weak Darboux equation for some metric $g$, but for which the auxiliary metric $h=g-du^2$ has nonzero distributional Gaussian curvature, or for which $h$ admits no $C^{1,\\theta}$ isometric immersion into $\\mathbb{R}^2$; either observation would disprove the converse direction of Theorem 1.1.","tokens_in":14457,"feed_emoji":"📐","tokens_out":8823,"duration_ms":95608,"temperature":0.7,"pith_summary":"The paper establishes that the classical correspondence between solutions of the Darboux equation and isometric immersions of surfaces into $\\mathbb{R}^3$ survives at low regularity. For any H\\\"older exponent $\\theta>1/2$, a function $u$ with $|du|_g<1$ satisfies Darboux's equation in a newly defined distributional sense exactly when it can be paired with some $\\phi$ so that $(\\phi,u)$ is a $C^{1,\\theta}$ isometric immersion of $(\\Omega,g)$. This matters because $C^{1,\\theta}$ is the borderline regime between the flexible and rigid behavior of isometric immersions, and weak constraint equations of this kind are the tools used to prove rigidity. The proof works by converting the Darboux equation into flatness of the auxiliary metric $h=g-du^2$, detected through a distributional Gaussian curvature defined for H\\\"older continuous metrics.","feed_headline":"Weak Darboux solutions are immersion heights for θ > 1/2","feed_subtitle":"A classical PDE-to-geometry bridge survives at low regularity, a tool for C^{1,θ} surface rigidity.","key_machinery":"The argument is carried by the auxiliary metric $h=g-du^2$. For a smooth $u$, the classical curvature formula says the flatness of $h$ is equivalent to the Darboux equation; the paper makes this equivalence work at low regularity by defining a distributional Gaussian curvature $K_h\\,dV_h$ for $C^{0,\\theta}$ metrics via conformal coordinates and distributional products, and proving a flatness criterion: such an $h$ admits a $C^{1,\\theta}$ isometric immersion into $\\mathbb{R}^2$ if and only if $K_h\\,dV_h=0$ as a distribution.","core_discovery":"The central claim is a two-way equivalence at the low-regularity threshold $\\theta>1/2$. If $r\\in C^{1,\\theta}$ is an isometric immersion of $(\\Omega,g)$ into $\\mathbb{R}^3$ and $u=r\\cdot e$ with $|du|_g<1$, then $u$ is a weak solution of the Darboux equation $\\det\\nabla^2_g u=K_g(1-|du|^2_g)$. Conversely, every $C^{1,\\theta}$ weak solution $u$ with $|du|_g<1$ can be completed by a $C^{1,\\theta}$ map $\\phi$ so that $r=(\\phi,u)$ is an isometric immersion of $(\\Omega,g)$ into $\\mathbb{R}^3$. The classical smooth correspondence therefore persists with exactly the same shape in the low-regularity regime, provided the equation is interpreted through distributional products and a distributional Ga","pith_inferences":["A natural testable extension: check whether the threshold $\\theta>1/2$ is sharp. If there is a $C^{1,1/2}$ weak Darboux solution that admits no $C^{1,1/2}$ isometric completion, then the theorem's exponent is optimal; the paper does not address this.","The stability statement in Lemma 4.1 suggests the Darboux-to-flat-metric construction is continuous under $C^{0,\\theta}$ metric perturbations, which could turn the correspondence into a compactness device for sequences of isometric immersions; this step is not taken in the paper.","The same mechanism of converting a Darboux-type equation into flatness of an auxiliary metric should extend to other Monge\\textendash{}Amp\\`ere type equations arising in codimension-one immersions, and to surfaces whose metrics have low H\\\"older regularity, since the proof uses only conformal coordinates and distributional products."],"forward_implications":["Every $C^{1,\\theta}$ isometric immersion of $(\\Omega,g)$ has each height component $u=r\\cdot e$ solving the Darboux equation in the distributional sense whenever $|du|_g<1$.","Every $C^{1,\\theta}$ weak Darboux solution can be completed by a $C^{1,\\theta}$ function $\\phi$ into an isometric immersion; weak Darboux solutions are exactly the height functions of such immersions.","Flat H\\\"older metrics of class $C^{0,\\theta}$ can be recognized by vanishing distributional Gaussian curvature, with no additional smoothness assumptions.","The correspondence is stable under mollification: approximations of a weak Darboux solution yield smooth metrics whose curvature distributions converge to the distributional curvature of $h=g-du^2$.","Rigidity questions for $C^{1,\\theta}$ surfaces can now be phrased as questions about uniqueness or existence of weak Darboux solutions, since the two-way correspondence identifies the two objects completely."],"supporting_citations":[{"why":"Supplies the classical curvature formula for the perturbed metric $h=g-du^2$ (Lemma 2.1) and the smooth correspondence between Darboux solutions and isometric immersions that the paper extends.","marker":"[18]"},{"why":"Supplies the Beltrami equation theory and Schauder estimates used in Appendix B to construct global $C^{1,\\theta}$ conformal charts with stability under metric perturbation.","marker":"[2]"},{"why":"Provides the variable-metric conformal mapping theorem used to ensure the global conformal charts are orientation preserving diffeomorphisms with positive Jacobian.","marker":"[1]"},{"why":"Cited as a known version of the distributional product of H\\\"older continuous functions that underlies both the weak Hessian determinant and the distributional Gaussian curvature.","marker":"[15]"},{"why":"Provides the Besov embedding used in Appendix A to prove the required continuous extension of the distributional product.","marker":"[33]"},{"why":"Original source of the Darboux equation that the paper's weak formulation generalizes.","marker":"[12]"},{"why":"Precedent for defining distributional Gaussian curvature at low regularity; the paper extends this from $C^1$ metrics to $C^{0,\\theta}$ metrics.","marker":"[32]"}],"fun_headline_variants":["Weak Darboux solutions equal immersion heights when θ > 1/2","Low-regularity Darboux solutions rebuild isometric immersions","Darboux weak solutions mirror isometric immersions for C^{1,θ}","θ > 1/2: Darboux weak solutions are immersion heights"],"cache_read_input_tokens":2816,"weakest_assumption_plain":"The proof stands on Lemma 4.1, which asserts that every $C^{0,\\theta}$ metric on a simply connected domain has a global conformal chart in $C^{1,\\theta}$ depending continuously on the metric; if that chart regularity or stability fails, the isometric immersion obtained from a weak Darboux solution could drop below $C^{1,\\theta}$.","fun_headline_variants_meta":{"raw":{"variants":["Weak Darboux solutions equal immersion heights when θ > 1/2","Low-regularity Darboux solutions rebuild isometric immersions","Darboux weak solutions mirror isometric immersions for C^{1,θ}","θ > 1/2: Darboux weak solutions are immersion heights"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000173,"raw_usage":{"total_tokens":1088,"prompt_tokens":691,"completion_tokens":397,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":435,"completion_tokens_details":{"reasoning_tokens":324}},"tokens_in":435,"tokens_out":397,"duration_ms":4920,"temperature":1.0,"reasoning_tokens":324,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-05T23:30:32.651076+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Construct a $C^{1,\\theta}$ function $u$ with $|du|_g<1$ that satisfies the weak Darboux equation for some metric $g$, but for which the auxiliary metric $h=g-du^2$ has nonzero distributional Gaussian curvature, or for which $h$ admits no $C^{1,\\theta}$ isometric immersion into $\\mathbb{R}^2$; either observation would disprove the converse direction of Theorem 1.1.","supporting_citations":[{"cited_title":"Isometric embedding of Riemannian manifolds in Euclidean spaces , volume 130 of Mathematical Surveys and Monographs","cited_arxiv_id":null,"evidence_quote":"Supplies the classical curvature formula for the perturbed metric $h=g-du^2$ (Lemma 2.1) and the smooth correspondence between Darboux solutions and isometric immersions that the paper extends."},{"cited_title":"Elliptic Partial Differential Equations and Quasicon- formal Mappings in the Plane (PMS-48)","cited_arxiv_id":null,"evidence_quote":"Supplies the Beltrami equation theory and Schauder estimates used in Appendix B to construct global $C^{1,\\theta}$ conformal charts with stability under metric perturbation."},{"cited_title":"Riemann’s mapping theorem for variable metrics","cited_arxiv_id":null,"evidence_quote":"Provides the variable-metric conformal mapping theorem used to ensure the global conformal charts are orientation preserving diffeomorphisms with positive Jacobian."},{"cited_title":"The geometry of c1,α flat isometric immersions","cited_arxiv_id":null,"evidence_quote":"Cited as a known version of the distributional product of H\\\"older continuous functions that underlies both the weak Hessian determinant and the distributional Gaussian curvature."},{"cited_title":"Theory of function spaces","cited_arxiv_id":null,"evidence_quote":"Provides the Besov embedding used in Appendix A to prove the required continuous extension of the distributional product."},{"cited_title":"Le¸ cons sur la th´ eorie g´ en´ erale des surfaces","cited_arxiv_id":null,"evidence_quote":"Original source of the Darboux equation that the paper's weak formulation generalizes."},{"cited_title":"Convexity of weakly regular surfaces of distributional nonnegative intrinsic curvature","cited_arxiv_id":null,"evidence_quote":"Precedent for defining distributional Gaussian curvature at low regularity; the paper extends this from $C^1$ metrics to $C^{0,\\theta}$ metrics."}],"review_version":1}