{"id":"5d1573f4-6797-45f0-bd88-481c87f2bd2e","arxiv_id":"2501.02106","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"A stress-field W^{1,2} regularity theorem is proved for weighted p-Laplace equations with log-concave weights that can degenerate at the boundary, globally on bounded convex domains and locally on unbounded ones.","lead":"This paper proves sharp second-order regularity estimates for solutions of weighted p-Laplace equations with log-concave weights that may vanish at the boundary, under Neumann boundary conditions on convex domains. It is a substantial advance in the theory of degenerate elliptic PDEs and includes the first such estimates for the linear case p=2.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Theorem 1.1 rests on the unproved weighted Poincaré inequality (3.1); the cited references must be checked to cover the nonlinear mean condition for all p>1.","rationale":"The reader identified the weighted Poincaré inequality (3.1) as the weakest assumption, and I agree: it is the only substantial external input in the proof of Theorem 1.1, and the theorem's estimates collapse without it. I checked the rest of the global argument for internal gaps. The passage to the limit in Step 2 uses a weak lower-semicontinuity argument that is not fully written (the displayed equality should be an inequality), but this is easily repaired. The application of (3.4) to A(∇u) before knowing A(∇u)∈L^2(Ω;ϱ) is formally circular, but it can be justified by applying (3.4) to truncations of A(∇u) and passing to the limit. The identification W=A(∇w) in Step 3 is stated via a.e. convergence that is not immediate from the displayed convergences, but it follows from the compactness of bounded sets in W^{1,2}(Ω;ϱ) combined with the continuity of A^{-1}. These are presentation issues, not fatal flaws. Therefore the only true soft spot is the unproved external inequality (3.1). Since the inequality is plausible and likely covered by the cited literature, I recommend keeping the reader's conditional verdict: the manuscript should be accepted only after (3.1) is either proved or precisely attributed to the references. This does not change the verdict, so verdict_should_be is UNCHANGED.","tokens_in":33512,"tokens_out":33090,"duration_ms":323763,"concrete_test":"Check the statements in [23] and [18] to determine whether they provide (3.1) verbatim for all p>1 with the nonlinear condition ∫|u|^{p-2}uϱ=0 and a constant C(p) independent of the weight. If they do not, supply a self-contained proof of (3.1) or replace it with a weaker inequality that still yields Lemma 5.1 and (5.39). As a numerical sanity check, compute the optimal constant in (3.1) for p=1.5, 2, 3 on the unit interval with weights x^a and e^{-x/ε}, and verify that the constant is bounded by C(p) d^p as a/ε vary.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The global estimate in Theorem 1.1 depends crucially on the weighted Poincaré inequality (3.1): it is used in Lemma 5.1 to obtain the energy estimate (5.2), in inequality (3.4) to control the L^2(Ω;ϱ)-norm of A(∇u) in (5.39), and in the compactness Theorem 3.2. The inequality is cited from [23,18] but not proved in the paper. The statement is nonstandard in two respects: it is required for every p>1, and the mean condition is the nonlinear one ∫|u|^{p-2}uϱ dx=0, not the usual linear weighted mean. If the cited results establish only p=2 or only the linear mean condition, then the proof of Lemma 5.1 and the final bound (1.16) do not go through as written. This is a verification concern rather than a suspected falsehood: log-concave measures on convex domains are generally known to satisfy Poincaré inequalities with diameter-type constants, so the inequality is plausible; however, the exact form used here is load-bearing and should be either proved or precisely located in the literature.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proves global and local second-order regularity for weighted p-Laplace Neumann problems with log-concave weights on convex domains. The main result, Theorem 1.1, shows that for a bounded convex domain Ω and a weight ρ=e^{-h} with h convex, any weak solution with compatible f∈L²(Ω;ρ)∩L^{p'}(Ω;ρ) has A(∇u)=|∇u|^{p-2}∇u ∈ W^{1,2}(Ω;ρ), together with the quantitative bounds (1.15) and (1.16). Theorems 1.2 and 1.3 give local boundary versions in unbounded convex domains, with separate statements for 1<p≤2 and for p>2 under the additional structural assumption ρ=g^a. The proof combines approximation of the domain, weight, and operator with a weighted Reilly identity, weighted Poincaré inequalities, and a compactness theorem for weighted Sobolev spaces.","tokens_in":33732,"tokens_out":39202,"duration_ms":399698,"significance":"If the technical gaps identified below are closed, the results are a meaningful extension of the known second-order regularity theory for the p-Laplace stress field to weights that are allowed to degenerate at the boundary; the p=2 case is already presented as new. The paper has several strengths: the approximation scheme is systematic, the constants in the main bounds are explicit, the weighted Reilly identity is stated quantitatively, and the limitation for Dirichlet problems is discussed honestly. The two points that need attention are the status of the nonlinear-mean weighted Poincaré inequality (3.1) and the missing strong-convergence argument in the final approximation steps; both are standard but load-bearing.","major_comments":[{"comment":"The weighted Poincaré inequality is stated as a known result from [23,18] in the nonstandard form with the nonlinear mean condition ∫|u|^{p-2}uρ=0 and for every p>1. This inequality is load-bearing: it is used in Lemma 5.1 for the energy estimate (5.2), in (3.4) for the final L² control (5.39), and in Theorem 3.2 for compactness. Please add a proof of (3.1) or a precise statement of the relevant theorem in [23,18]. The missing step is short: the constant c_u from Lemma 3.1 is the minimizer of c↦∫|u−c|^pρ, so ∫|u−c_u|^pρ ≤ ∫|u−(u)_{Ω;ρ}|^pρ, and the usual linear-mean Poincaré inequality gives (3.1).","section":"§3, Eq. (3.1)"},{"comment":"The displayed convergences give u_k⇀w weakly in W^{1,p} and u_k→w strongly in L^p_loc and a.e., but they do not imply A(∇u_k)→A(∇w) a.e., because a.e. convergence of u_k does not control ∇u_k. The identification W=A(∇w) therefore needs a separate argument. One should prove strong convergence of ∇u_k in L^p_loc by testing the difference of the equations for u_k and w with u_k−w and using the p-monotonicity of A together with f_k→f in L^{p'}; then A(∇u_k)→A(∇w) a.e. and the passage to the limit is justified.","section":"§5, Step 3, around (5.43)"},{"comment":"The final approximation step asserts 'u_k→u in W^{1,p}(Ω_R∩B_{R0}(x0);ρ)' and uses it to pass to the limit in (6.32), but no proof is supplied. Without strong convergence, only weak convergence is available and the limsup of ∫|A(∇u_k)|²ρ is not controlled by the corresponding quantity for u. The same monotonicity argument as in the previous comment should be written in both local proofs, since the estimates (1.21) and (1.25) depend on this passage.","section":"§6, Step 3 of Theorem 1.2 and Step 2 of Theorem 1.3"}],"minor_comments":[{"comment":"There are several typographical slips, for example 'satifies' in Theorem 1.1 and the definition of η_ε in §2.2, which should read η_ε(x)=ε^{-n}η(x/ε) rather than x/εn; please correct them throughout.","section":"Theorem 1.1 and §2.2"},{"comment":"In the Moser iteration display following (6.10), the factor q_k/q_k appears to be a typo for q_k^{k/q_k} or an equivalent exponent; please correct the displayed formula.","section":"Lemma 6.2"},{"comment":"The sentence 'After adding the quantity C1∫_σ ... to both sides' is followed by a display in which both sides are also divided by (1+C1); the text should state this division explicitly.","section":"Proof of Theorem 1.3, before (6.39)"},{"comment":"Reference [35] appears in the bibliography but does not seem to be cited in the body of the paper; please check whether it is used.","section":"References"},{"comment":"Theorem 1.3 allows a=0 in assumption (1.23), whereas Lemma 3.6 and its proof assume a>0; if the case a=0 (ρ≡1) is intended, add a sentence explaining that the lower bound (3.32) is trivial in that case.","section":"Theorem 1.3 and Lemma 3.6"}],"recommendation":"major_revision","confidential_remarks":"The central argument is structurally sound, and I found no reason to doubt the truth of the theorems. The revision should focus on the two load-bearing gaps described above; both have standard fixes. The stress-test concern about the nonlinear-mean Poincaré inequality is legitimate but can be resolved by the minimizer reduction, which should be stated explicitly rather than left to the reader."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper does something real: it carries the second-order Sobolev regularity program for the stress field A(∇u)=|∇u|^{p−2}∇u over to weights that vanish on the boundary of convex domains, and it covers p=2 as a new case. The weighted Reilly identity in Section 4 is clean, and the way the convexity of the domain and log-concavity of the weight give nonnegative boundary and Hessian terms is a nice mechanism. Theorem 1.1 is the flagship, and the constant in (1.15) being independent of the weight and domain is a sharp, attractive feature. The compactness theorem for weighted Sobolev spaces is a useful byproduct. I could not find a fatal error in the proof of Theorem 1.1, conditional on the weighted Poincaré inequality (3.1).\n\nThe soft spots are real but not disqualifying. The load-bearing input is (3.1): a Poincaré inequality for log-concave weights with the nonlinear mean condition ∫|u|^{p−2}uϱ=0, for every p>1. The paper cites [23,18] for this, but those references are about standard zero-mean Poincaré estimates; I am not convinced they cover the nonlinear mean condition in the stated generality. Since Lemma 5.1, (3.4), and the compactness theorem all use (3.1), Theorem 1.1 stands or falls on this. This is a verification issue, not a suspected falsehood, and it is probably fixable with a short proof or a precise reference. The authors should be asked to do that.\n\nThe local theorems 1.2–1.3 are less polished. Lemma 6.2 contains an explicitly admitted sketch of the Moser iteration for boundary boundedness, with the \"could not find a precise reference\" caveat. That is a gap in the exposition, not necessarily in the mathematics, but it needs to be expanded before the paper is rigorous. The extra structural assumption (1.23) for p>2 is handled honestly, and Remark 3.7 shows why it is needed.\n\nThe citation pattern is fine. Self-citations appear only where the authors' prior work is the relevant comparison, not as a substitute for the proof. No fitted parameters, no invented entities.\n\nWho is this for? People working on regularity for degenerate and quasilinear elliptic equations, weighted Sobolev spaces, and convex-domain boundary value problems. It deserves a serious referee: the main result is important, the method is reusable, and the gaps are concrete and likely repairable. I would send this to review and ask for a proof or precise reference for (3.1) and a complete Lemma 6.2.","headline":"A genuine extension of second-order regularity to degenerate log-concave weights, with the main theorem resting on a weight-independent Poincaré inequality that the authors cite but do not prove; worth refereeing.","tokens_in":796,"tokens_out":1147,"would_cite":true,"duration_ms":37849,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["35B65","35D30","35J25","35J62","35J70","35J92"],"pacs":[],"model":"deepseek-v4-flash","headline":"For weighted p-Laplace Neumann problems on convex domains, a log-concave weight guarantees the stress field |∇u|^{p−2}∇u lies in W^{1,2}(Ω;ϱ), with an explicit sharp estimate.","keywords":["degenerate elliptic equations","Neumann problems","boundary regularity","second-order derivatives","convex domains","log-concave weights","p-Laplace equation","Reilly identity"],"falsifier":"Compute the best constant in the weighted Poincaré inequality (3.1) for the log-concave weight $\\varrho(x)=e^{-1/x_n}$ on the unit cube $(0,1)^n$ and check whether it remains bounded by $C(p)\\,d_\\Omega^p$; the paper's proof of Lemma 5.1 and Theorem 1.1 requires that uniform bound. Equivalently, integrate the one-dimensional weighted Neumann problem with $\\varrho=x_n^a$ and a singular $f\\in L^2(\\Omega;\\varrho)$ of zero weighted mean; if the weighted Hessian integral diverges while the energy is finite, the theorem would be false.","tokens_in":33311,"feed_emoji":"📐","tokens_out":16029,"duration_ms":149092,"temperature":0.7,"pith_summary":"This paper proves that degeneracy in the weight does not destroy second-order regularity, provided the degeneracy is log-concave and the boundary condition is the natural zero-flux one. On any bounded convex domain, weak solutions to the weighted p-Laplace Neumann problem have their stress field $A(\\nabla u)=|\\nabla u|^{p-2}\\nabla u$ in the weighted Sobolev space $W^{1,2}(\\Omega;\\varrho)$, with an $L^2(\\varrho)$ estimate for $\\nabla A(\\nabla u)$ whose constant depends only on $n$ and $p$. This is new even in the linear case $p=2$, where it yields weighted $W^{2,2}$ regularity. The same mechanism gives local boundary regularity on unbounded convex domains, with a separate treatment for $p>2$ requiring weights of power type $\\varrho=g^a$. An independent byproduct is a compactness theorem for weighted Sobolev spaces under log-concave weights.","feed_headline":"Log-concave weights yield p-Laplace second-order regularity","feed_subtitle":"New global estimates hold even when the weight vanishes at the boundary, including p=2.","key_machinery":"The load-bearing object is a generalized Reilly identity for vector fields $V$ with $V\\cdot\\nu=0$ on $\\partial\\Omega$: $\\int_\\Omega \\varrho^{-1}(\\operatorname{div}(\\varrho V))^2\\,dx = \\int_\\Omega \\varrho\\,\\operatorname{tr}((\\nabla V)^2)\\,dx + \\int_{\\partial\\Omega}\\varrho\\,B(V_T,V_T)\\,dH^{n-1} + \\int_\\Omega \\varrho\\,\\nabla^2 h\\,V\\cdot V\\,dx$. Log-concavity makes $h=-\\log\\varrho$ convex, so the Hessian term is nonnegative; convexity of $\\Omega$ makes the boundary second-fundamental-form term nonnegative; and a matrix-ratio estimate gives $\\operatorname{tr}((\\nabla A_\\varepsilon(\\nabla u_\\varepsilon))^2)\\ge c(p)|\\nabla A_\\varepsilon(\\nabla u_\\varepsilon)|^2$. The proof then passes through three nested approximations (smooth convex domains, smooth log-concave weights, regularized stress fields $A_\\varepsilon$), with the weighted Poincaré inequality providing uniform energy bounds at each stage.","core_discovery":"For the weighted Neumann problem $-\\operatorname{div}(\\varrho\\,|\\nabla u|^{p-2}\\nabla u)=\\varrho f$ on a bounded convex domain $\\Omega$, the paper proves that whenever $\\varrho=e^{-h}$ is log-concave (possibly vanishing on $\\partial\\Omega$) and $f\\in L^2(\\Omega;\\varrho)\\cap L^{p'}(\\Omega;\\varrho)$ satisfies $\\int_\\Omega f\\varrho\\,dx=0$, every weak solution has the stress field $A(\\nabla u)=|\\nabla u|^{p-2}\\nabla u$ in $W^{1,2}(\\Omega;\\varrho)$. The main quantitative statement is $\\int_\\Omega |\\nabla A(\\nabla u)|^2\\varrho\\,dx\\le C_0(n,p)\\int_\\Omega f^2\\varrho\\,dx$, with the $L^2(\\Omega;\\varrho)$-norm of $A(\\nabla u)$ controlled by $f^2$ and $|f|^{p'}$; the constant $C_0$ is sharp, reducing to the classical sharp constant $1$ for $p=2$, $\\varrho=1$. For unbounded convex domains, local boundary regularity is proved for $1<p\\le 2$, and for $p>2$ under the additional structural assumption $\\varrho=g^a$ with $g$ concave.","pith_inferences":["The dependence on the cited weighted Poincaré inequality suggests the global theorem should survive for any weight class for which that inequality holds with a uniform constant; log-concavity is one sufficient condition rather than the only one.","The restriction $\\varrho=g^a$ in the $p>2$ local theorem is tied to a lower bound on the $\\varrho$-mass of boundary rectangles; weights like $\\exp(-1/x_n)$ violate that bound, indicating a genuine threshold for the local technique.","At $p=2$ the result can be read as a degenerate-mass version of classical Neumann $W^{2,2}$ theory, with the log-concave weight playing the role of an invariant density; this may be useful for quantitative error estimates in Fokker-Planck type problems."],"forward_implications":["For bounded convex domains, every weak Neumann solution with data in $L^2(\\Omega;\\varrho)\\cap L^{p'}(\\Omega;\\varrho)$ automatically has a weighted second-order derivative of the stress field, so no boundary smoothness beyond convexity is needed.","The constant in the gradient estimate is scale-invariant and, in the linear constant-weight case, recovers the optimal constant $1$ of the Neumann Poisson estimate.","For unbounded convex domains, the local boundary estimates give the same regularity up to the boundary for $1<p\\le 2$, and for $p>2$ whenever $\\varrho=g^a$.","The compactness result for $W^{1,p}(\\Omega;\\varrho)$ follows from the same weighted Poincaré inequality and gives a tool for studying degenerate quasilinear problems beyond the present equation."],"supporting_citations":[{"why":"Supplies the weighted Poincaré inequality for log-concave weights on convex domains used in Lemma 5.1 and the compactness theorem.","marker":"[23]"},{"why":"Sharp weighted anisotropic Poincaré inequality for convex domains; together with [23] it grounds estimate (3.1).","marker":"[18]"},{"why":"Establishes the unweighted p-Laplace global second-order estimates that the weighted result extends.","marker":"[14]"},{"why":"Recent global second-order regularity for p-Laplacian type equations with variable coefficients, the comparison baseline for the weighted theorem.","marker":"[38]"},{"why":"Provides global second-order estimates in anisotropic problems and the Reilly-type framework adapted here.","marker":"[2]"},{"why":"Gives uniform ellipticity and stress-field regularity in divergence-form equations; supplies the algebraic matrix-ratio inequality (2.13).","marker":"[27]"},{"why":"Log-concavity preservation under convolution, used to construct smooth log-concave approximations of the weight.","marker":"[42]"},{"why":"Boundary $C^{1,\\gamma}$ regularity for degenerate elliptic equations used to control the approximating solutions.","marker":"[36]"}],"fun_headline_variants":["Sharp p-Laplace estimates for log-concave weights","Even p=2: second-order regularity for weighted p-Laplace","Degenerate log-concave weights yield sharp p-Laplace bounds","New global estimates for weighted p-Laplace with vanishing weights"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The global proof inherits the weighted Poincaré inequality (3.1) for log-concave weights on bounded convex domains from cited references, without proving it; if that inequality failed, the uniform energy estimate and the final control of $A(\\nabla u)$ in $L^2(\\Omega;\\varrho)$ would not follow.","fun_headline_variants_meta":{"raw":{"variants":["Sharp p-Laplace estimates for log-concave weights","Even p=2: second-order regularity for weighted p-Laplace","Degenerate log-concave weights yield sharp p-Laplace bounds","New global estimates for weighted p-Laplace with vanishing weights"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000237,"raw_usage":{"total_tokens":1481,"prompt_tokens":891,"completion_tokens":590,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":507,"completion_tokens_details":{"reasoning_tokens":519}},"tokens_in":507,"tokens_out":590,"duration_ms":6173,"temperature":1.0,"reasoning_tokens":519,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T22:16:08.110713+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the best constant in the weighted Poincaré inequality (3.1) for the log-concave weight $\\varrho(x)=e^{-1/x_n}$ on the unit cube $(0,1)^n$ and check whether it remains bounded by $C(p)\\,d_\\Omega^p$; the paper's proof of Lemma 5.1 and Theorem 1.1 requires that uniform bound. Equivalently, integrate the one-dimensional weighted Neumann problem with $\\varrho=x_n^a$ and a singular $f\\in L^2(\\Omega;\\varrho)$ of zero weighted mean; if the weighted Hessian integral diverges while the energy is finite, the theorem would be false.","supporting_citations":[{"cited_title":"Ferone, C","cited_arxiv_id":null,"evidence_quote":"Supplies the weighted Poincaré inequality for log-concave weights on convex domains used in Lemma 5.1 and the compactness theorem."},{"cited_title":"Della Pietra, G","cited_arxiv_id":null,"evidence_quote":"Sharp weighted anisotropic Poincaré inequality for convex domains; together with [23] it grounds estimate (3.1)."},{"cited_title":"Cianchi, V.G","cited_arxiv_id":null,"evidence_quote":"Establishes the unweighted p-Laplace global second-order estimates that the weighted result extends."},{"cited_title":"Miao, Fa Peng, Y","cited_arxiv_id":null,"evidence_quote":"Recent global second-order regularity for p-Laplacian type equations with variable coefficients, the comparison baseline for the weighted theorem."},{"cited_title":"Guarnotta, S","cited_arxiv_id":null,"evidence_quote":"Gives uniform ellipticity and stress-field regularity in divergence-form equations; supplies the algebraic matrix-ratio inequality (2.13)."},{"cited_title":"Pr´ ekopa,On logarithmic concave measures and functions , Acta Sci","cited_arxiv_id":null,"evidence_quote":"Log-concavity preservation under convolution, used to construct smooth log-concave approximations of the weight."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Boundary $C^{1,\\gamma}$ regularity for degenerate elliptic equations used to control the approximating solutions."}],"review_version":1}