{"id":"3c124493-4faa-409e-b3d3-4941faa5c1d1","arxiv_id":"2412.11902","paper_version":2,"verdict":"ACCEPT","confidence":"MODERATE","novelty_score":8.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"For dimensions n ≤ 4, any volume m > 0, and any admissible periodic coefficients, an overdetermined semilinear boundary problem admits a positive smooth solution on a smooth bounded domain.","lead":"Mathematicians prove that for every smooth periodic and admissible semilinear elliptic problem, there exists a smooth bounded domain of any prescribed size where a positive solution exists together with the required normal derivative at the boundary. This removes the free-parameter trick used in all earlier constructions and is new even for the Poisson equation with a position-dependent source and constant Neumann data.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Step 3 of §8 shows only that the translated function minimizes the truncated energy F0,R; the conclusion that it minimizes the original F0 does not follow from the displayed inequalities.","rationale":"The paper's core variational machinery is substantial and mostly coherent: the coercivity estimate, the Lipschitz and non-degeneracy results, the free-boundary analysis, and the use of periodicity to control supports are all plausible and carefully connected. The reader's identification of x-periodicity and the smallness condition b<λλ1(Bm)/2 as fragile inputs is reasonable. However, the most load-bearing gap I found is different: Section 8 proves that the constructed function minimizes the truncated functional F0,R, and then asserts all the properties of Theorem 1.2 without proving that it minimizes the original F0. The displayed inequalities point in the wrong direction for such a conclusion, and the periodic translation argument is applied only to the minimizer, not to arbitrary competitors. This is a genuine missing step in the proof of the paper's stated minimizer theorem. The gap appears repairable by a standard diagonal compactness argument using the uniform bounds already established, so it does not warrant rejection, but it requires a substantive addition rather than a cosmetic clarification. Therefore I recommend a conditional acceptance pending the missing argument.","tokens_in":45307,"tokens_out":32961,"duration_ms":316621,"concrete_test":"Check the final step of §8 against an explicit competitor v∈K≤m whose support is a union of many small components spread far outside BR, so that F0,R(v)=∫|∇v|2 while F0(v)=∫|∇v|2−2∫F(x,v). Since Step 3 only yields F0,R(\\tilde u0)≤F0,R(v), the displayed estimates cannot rule out F0(v)<F0(\\tilde u0). Then attempt the missing compactness argument: take minimizers u_R of F0,R along a sequence R→∞, use the uniform H1 and support bounds to extract u_R→u in H1, and verify that lower semicontinuity gives F0(u)≤liminf F0(u_R), which would complete the proof of global minimality. If this argument cannot be carried out, Theorem 1.2's minimizer claim is not proven.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"In §8 the authors minimize the truncated functional F0,R, produce a minimizer u0, and then translate its enlarged connected components into a fixed ball BR, obtaining \\tilde u0 with F0,R(\\tilde u0)=inf F0,R and FR=F on Ω_{\\tilde u0}. The final sentence concludes that u:=\\tilde u0 has all the properties listed in Theorem 1.2, including being a minimizer of the original F0 in K≤m. This is not established by the written proof. For an arbitrary v∈K≤m, since F≥0 and φR≤1, we have F0(v)≤F0,R(v), so the inequality F0,R(\\tilde u0)≤F0,R(v) only gives F0(\\tilde u0)≤F0,R(v); it cannot be compared with F0(v). The periodic-translation device is applied to the minimizer, not to arbitrary competitors, and Proposition 6.2 provides no finite-component or diameter bound for general elements of K≤m. A missing compactness argument taking R→∞ (using the uniform H1 and support bounds to extract a subsequential limit) would be needed to obtain a genuine minimizer of F0. Without that step, Theorem 1.2's existence-of-minimizer assertion is unsupported as written, although the PDE/free-boundary conclusions may still be recoverable from truncated minimality.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proves existence of bounded open sets with prescribed q-volume supporting positive solutions to the overdetermined semilinear elliptic problem (1.2) with position-dependent, periodic coefficients, in dimensions n <= 4, with a codimension-5 bound on the singular set in higher dimensions. The method is variational: one minimizes a one-phase Alt-Caffarelli energy over H^1(R^n) under a volume constraint, derives Lipschitz regularity, non-degeneracy, boundedness of the support, free-boundary regularity and the overdetermined Neumann condition for an assumed minimizer, and then constructs minimizers for truncated periodic energies in Section 8. A compact-manifold version and an appendix showing the necessity of periodicity for uniform support bounds are included.","tokens_in":45549,"tokens_out":9468,"duration_ms":85975,"significance":"If correct, this is a substantial advance: it supplies a non-perturbative existence theory for a general class of overdetermined problems without a free parameter, and is new even for the Poisson equation with constant Neumann data. The proof is detailed and builds on the Alt-Caffarelli program, Velichkov's monograph, and the De Silva-Ferrari-Salsa regularity theory. The paper is unusually honest about its fragile inputs: it flags the non-positivity of the Weiss derivative, repairs the initial non-negativity of the Lagrange multiplier, and shows in Appendix A that periodicity is genuinely needed for uniform support control. These strengths make the result worth publishing after the Section 8 gaps described below are repaired.","major_comments":[{"comment":"The truncated energy is defined as F0,R(v,D) := ∫_D |∇v|^2 dx − 2∫_D FR(x,v) dx, without the coefficient matrix A(x) that appears in the original energy (1.5). Step 2 then invokes Corollary 5.4 and Proposition 7.9, whose variational and Euler-Lagrange structure is built on the operator L = −div(A∇·). With the displayed definition, the minimizer would satisfy −Δu = F_R'(x,u), not the equation in (1.6). Please replace |∇v|^2 by ∇v A∇v^T, or explicitly justify a reduction to A ≡ I, and recheck (8.2) and the lower-semicontinuity estimates accordingly. As written, the construction in Step 1 is for a different functional.","section":"§8, Step 1"},{"comment":"The final conclusion 'so is ũ0' and 'u := ũ0 has all the properties listed in Theorem 1.2' includes the assertion that u minimizes the original functional F0. The displayed argument only proves F0,R(ũ0) = inf F0,R and FR ≡ F on Ω_{ũ0}. For an arbitrary v ∈ K≤m one has F0(v) ≤ F0,R(v) (since F ≥ 0 and φR ≤ 1), while the inequality obtained is F0(ũ0) ≤ F0,R(ũ0) ≤ F0,R(v); this does not imply F0(ũ0) ≤ F0(v). The translation device applies to the constructed minimizer, not to arbitrary competitors, and Proposition 6.2 provides no diameter bound for general elements of K≤m. A limiting argument R→∞, using the uniform H^1 and support bounds to extract a subsequential limit, is needed to obtain a genuine minimizer of F0. Without that step, Theorem 1.2's existence-of-minimizer assertion is unsupported as written, although the PDE and free-boundary conclusions may still be recoverable from truncated minimality.","section":"§8, Step 3"},{"comment":"The step chooses R large enough after translating the enlarged connected components, relying on the assertion that N and D are independent of R. However, Propositions 5.3, 5.6 and 6.2 produce constants L, κ0, r0, N, D that depend on the Lagrange multiplier Λ from Lemma 4.1, and Lemma 4.1 only guarantees the existence of some Λ ≥ 0 for each minimizer; no uniform bound for Λ_R, and hence for L_R, κ0,R, is proved. Unless such uniformity is established, the constants N,D used to fix R may themselves depend on R, making the choice of R circular. Please provide a uniformity argument or an alternative construction that avoids this dependence.","section":"§8, Step 3"}],"minor_comments":[{"comment":"The sentence 'Our main result can the be stated' should read 'can then be stated'.","section":"§1.1"},{"comment":"The phrase 'even with he admissibility and periodicity assumptions' contains a typo: 'he' should be 'the'.","section":"§3"},{"comment":"The reference [DEP1] is listed twice with different titles; the second entry, for the Arch. Ration. Mech. Anal. 239 (2021) paper on stationary Euler flows, should be given a distinct key such as [DEP2] or a new label.","section":"References"},{"comment":"The caption refers to 'cubes of length K', while the text uses 2D and 2DN′; please align the notation.","section":"Figure 3"},{"comment":"In Proposition 7.9, Step 1, the passage from u0 to bu0 and back is implicit when deriving the bound on |∇φ(0)|; a brief sentence clarifying that the displayed inequality is after undoing the change of variables would help the reader.","section":"§7.2"}],"recommendation":"major_revision","confidential_remarks":"The paper is strong and the variational strategy is original, but Section 8 contains two genuine gaps: the truncated energy as written omits A, and the passage from truncated to original minimality is not justified; a third point about the uniformity of the constants in R also needs attention. All three appear fixable within the scope of the manuscript, so I recommend major revision rather than rejection. There is no concern about attribution or overlap beyond the normal."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper does something genuinely new: for n≤4 it produces smooth domains of arbitrary prescribed q-volume on which a position-dependent semilinear overdetermined problem admits a positive solution, with no free parameter λ. This is new even for the Poisson equation with constant Neumann data. The variational program on the whole space, with nonnegative nonlinearity, is the right idea, and the authors work hard for it: uniform L∞ and Lipschitz bounds, non-degeneracy, a Weiss-type monotonicity argument with only a lower bound on the derivative, and free-boundary regularity via De Silva–Ferrari–Salsa. Appendix A is honest and useful, showing why periodicity is not merely technical. The citation pattern is solid and there is no circularity: they minimize an energy, not fit a target.\n\nThe soft spot is real and it is in Section 8, Step 3. The authors minimize the truncated energy F0,R, obtain u0, translate its enlarged components into a fixed ball, and then claim the translated function has all the properties listed in Theorem 1.2, including being a minimizer of the original F0. That final conclusion does not follow. For an arbitrary competitor v, F0(v) ≤ F0,R(v) because F ≥ 0 and φR ≤ 1. So the displayed inequality F0,R(ũ0) ≤ F0,R(v) gives F0(ũ0) ≤ F0,R(v), which cannot be compared with F0(v). The step from truncated to untruncated minimization needs a compactness argument taking R→∞, using the uniform H1 and support bounds; that argument is not in the paper. This is not a cosmetic detail, and the referee should require it.\n\nThat said, the main PDE result is probably salvageable. On the support of the translated function, FR = F, so the local Euler–Lagrange equation, the free-boundary condition, and the regularity analysis all go through. Thus Theorem 1.1, the existence of smooth domains and positive solutions, appears well supported even if the global-minimizer assertion in Theorem 1.2 is withdrawn or proved later. The paper is significant, mostly careful, and the gap is identifiable and plausibly fixable. It deserves a serious referee, not a desk rejection. I would send it out, with a pointed request to address Section 8 Step 3.","headline":"Genuinely new existence theorem for overdetermined semilinear problems, well worth refereeing, but Section 8 Step 3 has a real gap between truncated and original minimization.","tokens_in":46074,"tokens_out":3315,"would_cite":true,"duration_ms":32491,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["35R35","35J25","35J61","35B65"],"pacs":[],"model":"deepseek-v4-flash","headline":"For dimensions up to four, any smooth periodic admissible choice of coefficients and nonlinearity, together with any prescribed q-weighted volume, yields a bounded domain with smooth boundary on which the semilinear overdetermined problem…","keywords":["overdetermined boundary problems","semilinear elliptic equations","free boundary regularity","one-phase variational problems","compact support of minimizers","periodic coefficients","Neumann boundary condition","singular set dimension"],"falsifier":"Compute the energy of a first-eigenfunction spike for the supercritical quadratic nonlinearity: with $F(x,u)=b u^2$, $A=I$, $q=1$ and $b>\\lambda_1(B_m)/2$, one gets $F_0(\\tau\\varphi_1)=\\tau^2(\\lambda_1(B_m)-2b)\\int\\varphi_1^2\\to -\\infty$, so no minimizer exists and the growth assumption (HF4) is sharp. Independently, the Appendix A two-bump example shows that dropping periodicity makes the support diameter of minimizers arbitrarily large, contradicting the uniform support bound on which Theorem 1.2 depends.","tokens_in":45093,"feed_emoji":"📐","tokens_out":16972,"duration_ms":142743,"temperature":0.7,"pith_summary":"This paper establishes an existence theorem for a general class of semilinear elliptic overdetermined boundary problems: for $n\\leq 4$, given smooth, $x$-periodic, admissible coefficient data $A,q$ and nonlinearity $f$, and any $m>0$, there is a bounded open set $\\Omega\\subset\\mathbb{R}^n$ with smooth boundary, $q$-weighted volume $m$, and a constant $c>0$ such that the problem $-\\operatorname{div}(A\\nabla v)=f(x,v)$ in $\\Omega$, $v=0$ and $\\nabla v A\\nabla v^T=c\\,q$ on $\\partial\\Omega$, admits a positive solution $v\\in C^{\\infty}(\\Omega)$. Here admissibility roughly means $A$ is uniformly elliptic, $q$ is bounded above and below, and $f$ is nonnegative and grows at most linearly. The proof treats the unknown domain as the support of a minimizer of a one-phase free-boundary energy on the whole space, and the main technical achievement is controlling the minimizer on the unbounded setting: uniform $L^{\\infty}$ bounds, bounded support, and uniform bounds on the number and diameter of the support's components. The method also gives free-boundary regularity: outside a singular set of Hausdorff dimension at most $n-5$ (empty for $n\\leq 4$), $\\partial\\Omega$ is $C^{1,\\alpha}$ and smooth when the data are smooth; the results are new even for the Poisson equation $-\\Delta v=g(x)$ with constant Neumann data.","feed_headline":"For n≤4, every admissible overdetermined problem has a solution","feed_subtitle":"For n≤4 and any prescribed q-volume, the overdetermined problem admits a positive smooth solution.","key_machinery":"The load-bearing object is the variational problem for $F_0(u)=\\int_{\\mathbb{R}^n}\\big(\\nabla u A\\nabla u^T-2F(x,u)\\big)\\,dx$ over the class $K_{\\leq m}=\\{u\\in H^1(\\mathbb{R}^n): u\\geq 0,\\ \\operatorname{Vol}_q(\\{u>0\\})\\leq m\\}$, with $F$ the primitive of $f$ and the overdetermined constant $c$ emerging as a Lagrange multiplier $\\Lambda$. The proof is a complete one-phase free-boundary program on the unbounded space: an approximate mean-value formula for inhomogeneous divergence-form equations, uniform boundedness of minimizers through subharmonic comparison with Newtonian potentials, Lipschitz regularity and non-degeneracy from comparison with harmonic replacements, compact support from a corkscrew condition, and a modified almost-monotone scaling identity whose derivative is only bounded below, enough to prove that blow-up limits are $1$-homogeneous. Those limits are then classified, forcing the viscosity Neumann condition $\\nabla u A\\nabla u^T=\\Lambda q$; a regularity theorem for divergence-form free boundaries then gives the $C^{1,\\alpha}$ regular set and the $n-5$ singular-set bound. Periodicity enters in the final step: it lets each enlarged connected component of the support be translated by whole periods into a fixed large ball without increasing the energy, which gives uniform support bounds and lets the truncated nonlinearity be replaced by the original $F$.","core_discovery":"The central claim is Theorem 1.2: for any $m>0$ and admissible $x$-periodic data, the energy $F_0$ has a Lipschitz minimizer $u$ in $K_{\\leq m}$ that saturates the volume constraint, $\\operatorname{Vol}_q(\\{u>0\\})=m$; the set $\\Omega=\\{u>0\\}$ is open and bounded, and in $\\Omega$ the minimizer solves $-\\operatorname{div}(A\\nabla u)=F'(x,u)$, while on $\\partial\\Omega$ it satisfies $\\nabla u A\\nabla u^T=\\Lambda q$ in the viscosity sense for some positive constant $\\Lambda$, which is the $c$ of the overdetermined problem. The free boundary splits into a regular part $\\operatorname{Reg}(\\partial\\Omega)$, a $C^{1,\\alpha}$ hypersurface open in $\\partial\\Omega$ on which the Neumann condition holds classically and which is smooth when $A,q,f$ are smooth, and a singular part $\\operatorname{Sing}(\\partial\\Omega)$ of Hausdorff dimension at most $n-5$, empty for $n\\leq 4$ and at most countable for $n=5$. Taking $v=u|_{\\Omega}$ gives Theorem 1.1: in dimensions $n\\leq 4$ the boundary is smooth, so the overdetermined problem has a positive $C^{\\infty}$ solution on a bounded domain of any prescribed $q$-volume. A key point of the paper is that the construction uses no free eigenvalue parameter and does not start from a small perturbation of a ball.","pith_inferences":["The $n\\leq 4$ restriction probably comes from the standing $n-5$ bound on the singular set: if that bound were improved, the identical construction would yield smooth domains in higher dimensions, while a minimizer in $n=5$ with an isolated conical boundary point would show the threshold is sharp.","Periodicity acts as a compactness normalization: the Appendix A two-bump example suggests that without it minimizers can be translated arbitrarily far apart, so a non-periodic theory would have to add a selection mechanism, such as pinning the support or working in a quotient.","In the constant-coefficient torsion case ($A=I$, $q=1$, $f\\equiv 1$), the classical symmetry theorem forces the domain to be a ball; comparing the Lagrange constant $c$ with the explicit value $|\\nabla u|=R/n$ on that ball would be a direct check of the free-boundary machinery.","Because the proof only needs the almost-monotone scaling identity to have derivative bounded below, one could try to weaken the smoothness of $A,q,f$ from $C^{1,1}$ to H\\\"older classes and still obtain Lipschitz minimizers with the same free-boundary structure."],"forward_implications":["For $n\\leq 4$ and any $m>0$, the construction yields a bounded domain with smooth boundary and $q$-volume $m$ on which the overdetermined problem (1.2) has a positive smooth solution; no eigenvalue parameter is needed.","The free boundary of the minimizer is mostly smooth: a regular part that is $C^{1,\\alpha}$ (smooth for smooth data) and a singular part of Hausdorff dimension at most $n-5$, empty for $n\\leq 4$ and at most countable for $n=5$.","The minimizer always saturates the constraint, $\\operatorname{Vol}_q(\\{u>0\\})=m$, so the full prescribed mass is used.","The same argument proves the Laplace--Beltrami analogue on compact Riemannian manifolds of dimension $n\\leq 4$ for any $0<m<\\operatorname{Vol}_{q,g}(M)$.","The theorem covers the Poisson equation $-\\Delta v=g(x)$ with constant Neumann data, a case not previously handled."],"supporting_citations":[{"why":"supplies the variational one-phase free-boundary machinery (energy, Lipschitz and non-degeneracy estimates, free-boundary regularity) that the paper adapts to the whole space.","marker":"[AC]"},{"why":"supplies the modern blueprint for asymptotic minimality, scaling identities, and the dimension bound for the singular set that Sections 4–7 follow.","marker":"[V]"},{"why":"supplies the interior and boundary elliptic estimates used repeatedly for regularity of minimizers and of the free boundary.","marker":"[GT]"},{"why":"supplies comparability of Green functions for divergence-form operators, used in the uniform $L^{\\infty}$ bound and in the mean-value estimates.","marker":"[LSW]"},{"why":"supplies the boundedness of Riesz potentials that drives the self-improvement of integrability in the proof of uniform boundedness.","marker":"[St]"},{"why":"supplies the elliptic-measure comparison estimate needed for the approximate Poisson formula for inhomogeneous equations.","marker":"[CFMS]"},{"why":"supplies the $C^{1,\\alpha}$ free-boundary regularity theorem for divergence-form equations used to prove the regular part is a manifold.","marker":"[DFS]"},{"why":"supplies existence of volume-constrained Faber–Krahn minimizers on compact manifolds, used for the manifold version in Section 9.","marker":"[LS]"}],"fun_headline_variants":["Every admissible overdetermined problem has a positive solution for n≤4","Positive solutions for overdetermined problems on any volume when n≤4","Overdetermined boundary problems with positive solutions in dimensions ≤4","For n≤4, every volume admits an overdetermined positive solution","No free eigenvalue: positive solutions on any volume for n≤4"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing hypothesis is $x$-periodicity of $A$, $q$ and $F$: it is what allows each enlarged connected component of a minimizer's support to be translated into one fixed large ball without increasing the energy, and Appendix A shows that without periodicity admissible problems can have minimizers whose support diameter is arbitrarily large; the quadratic growth of $F$ must also stay below the coercivity threshold $b<\\lambda\\lambda_1(B_m)/2$.","fun_headline_variants_meta":{"raw":{"variants":["Every admissible overdetermined problem has a positive solution for n≤4","Positive solutions for overdetermined problems on any volume when n≤4","Overdetermined boundary problems with positive solutions in dimensions ≤4","For n≤4, every volume admits an overdetermined positive solution","No free eigenvalue: positive solutions on any volume for n≤4"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001397,"raw_usage":{"total_tokens":5703,"prompt_tokens":1054,"completion_tokens":4649,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":670,"completion_tokens_details":{"reasoning_tokens":4556}},"tokens_in":670,"tokens_out":4649,"duration_ms":30281,"temperature":1.0,"reasoning_tokens":4556,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T14:27:47.510359+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the energy of a first-eigenfunction spike for the supercritical quadratic nonlinearity: with $F(x,u)=b u^2$, $A=I$, $q=1$ and $b>\\lambda_1(B_m)/2$, one gets $F_0(\\tau\\varphi_1)=\\tau^2(\\lambda_1(B_m)-2b)\\int\\varphi_1^2\\to -\\infty$, so no minimizer exists and the growth assumption (HF4) is sharp. Independently, the Appendix A two-bump example shows that dropping periodicity makes the support diameter of minimizers arbitrarily large, contradicting the uniform support bound on which Theorem 1.2 depends.","supporting_citations":[],"review_version":1}