{"id":"a2f25baf-7a45-4d2d-9bfc-4feca79b4968","arxiv_id":"2507.11837","paper_version":1,"verdict":"ACCEPT","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"low","formal_verification":"none","parameter_count":1,"one_line_summary":"The authors construct case (c) least-total-curvature steady Euler flows in a strip and stable monotone semilinear solutions with non-convex superlevel sets.","lead":"This paper constructs new families of steady fluid flows in an infinite strip, completing the existence picture for least total curvature Euler flows. It also builds stable positive solutions of a semilinear elliptic equation whose level sets are not convex, answering a generalized question posed by Hamel, Nadirashvili and Sire.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"No significant objection identified: the heteroclinic construction and both existence theorems are sound; the only delicate point, Assumption 2.1, is verified in detail.","rationale":"The reader correctly identified Assumption 2.1 as the technical heart of the paper, and I agree that this is where the argument is most delicate. However, my stress-test of that assumption found it adequately supported: Proposition 3.1 supplies the two ordered minimizers, proves total ordering of the minimizer set, and constructs a minimal minimizer above phi; Proposition 4.1 reproduces the same structure for the c=0 setting. The heteroclinic construction in Proposition 2.1 is standard and internally consistent, including the Hamiltonian identity, the reference-point normalization, and the Hopf-lemma contradiction that rules out boundary loss of the transition. Theorem 1.3's non-convexity proof is logically tight: the superlevel set contains a full horizontal line, convexity forces it to be a strip, and the asymptotic limits then force phi and phi to agree at two interior points, contradicting their strict ordering. The stability claim follows from the positive x1-derivative via the standard identity Q(xi) = integral of w^2 |nabla eta|^2, with w = partial_x1 u > 0 and eta = xi/w. I found no internal inconsistency, unsupported essential assumption, or circular step. The remaining issues are cosmetic typos and an unstated but elementary continuity argument for m_lambda; these do not affect the central claims and do not warrant changing the reader's ACCEPT verdict.","tokens_in":16603,"tokens_out":41457,"duration_ms":496953,"concrete_test":"Verify Assumption 2.1 numerically for the constructed F_lambda: compute the 1D minimizers at lambda* by shooting for -psi'' = f_lambda*(psi) with the relevant boundary conditions, and check that the only global minimizers are phi and phi, with equal energies and no third minimizer strictly between them.","verdict_should_be":"UNCHANGED","load_bearing_attack":"I read the paper in good faith and could not identify a load-bearing flaw. The central claim rests on Assumption 2.1: two ordered global minimizers of the 1D energy with no minimizer between. Proposition 2.1 builds the heteroclinic through standard truncation, monotonicity from the principal eigenvalue, and a Hamiltonian/limit argument; Lemma 2.2 correctly prevents loss of the transition. Propositions 3.1 and 4.1 construct F_lambda so that t (resp. t(1-t)) is a nondegenerate local minimizer, the set E of lambda with energy below the trivial value is nonempty and bounded, and at lambda* a second minimizer above the trivial one is obtained by compactness and total ordering of the minimizer set. The total-ordering/min-max argument is valid, and the minimal-minimizer construction rules out minimizers strictly between phi and phi. Theorem 1.3's convexity contradiction is correct: if the superlevel set containing the line R x {1/2} were convex, it would be a full strip R x (a,b), forcing phi(a)=phi(a)=alpha, contradicting phi<phi. Minor gaps, such as the unstated continuity of m_lambda needed to justify m_{lambda*}=1/2, are routine and fixable, not load-bearing. Typos and the mislabeled internal step references do not affect the arguments.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies steady incompressible Euler flows in the strip Ω∞ = R × (0,1) that belong to case (III) of the Hamel–Nadirashvili classification, the so-called least total curvature flows. It constructs heteroclinic solutions to a semilinear elliptic equation with prescribed boundary values, assuming two ordered global minimizers of the associated one-dimensional energy with no minimizer between them (Assumption 2.1), following and extending the framework of [16]. The authors then engineer nonlinearities so that Assumption 2.1 holds. For Theorem 1.2, connecting an increasing minimizer to a concave non-monotone minimizer yields a bounded smooth Euler flow with v2 > 0 and boundary signs corresponding to the previously missing case (c). For Theorem 1.3, two positive symmetric minimizers with c = 0 yield a positive, x1-monotone (hence stable) solution with a nonconvex superlevel set, giving a negative answer to the generalized quasiconcavity-from-semi-stability question in [27] in the unbounded convex strip. The proofs combine variational minimization on truncated boxes, principal-eigenvalue maximum principles, Hamiltonian identities, and explicit cutoff nonlinearities.","tokens_in":16924,"tokens_out":13011,"duration_ms":153265,"significance":"If the results hold, they complete the existence picture for least-total-curvature flows in a strip by treating the missing case (c), complementing the constructions in [23] and [16]. The more striking contribution is Theorem 1.3: it shows that in an unbounded convex domain, positivity and monotonicity (which imply stability) do not force convex superlevel sets for semilinear elliptic equations, answering a generalized version of an open problem in [27] negatively. The paper is self-contained in its verification of Assumption 2.1, and the variational and heteroclinic arguments are standard and rigorous. No machine-checked code or numerical data is involved, but the constructions are sufficiently explicit that the key identities—energy comparisons, Hamiltonian conservation, nondegeneracy of the trivial minimizer—can be checked by hand.","major_comments":[],"minor_comments":[{"comment":"The assertion that mλ* = 1/2 follows 'by the definition of λ*' is not immediate from the monotonicity of mλ alone; one should add the standard argument that if mλ* < 1/2 with a minimizer φ*, then using φ* as a test function for λ > λ* close to λ* gives mλ < 1/2, contradicting the definition of λ*. The same remark applies to the hat-functional in Proposition 4.1.","section":"Section 3, Step 4 and Section 4, Step 4"},{"comment":"In the proof of Lemma 2.2, the line 'By maximum principle, ∂x1w < 0' should read ∂x1w > 0, because w is a limit of functions that are strictly increasing in x1 and has nonnegative boundary data; the subsequent Hopf-lemma sentence also appears to contain a sign typo and an undefined symbol v (probably w̃). These are local typos and do not affect the contradiction argument.","section":"Section 2, Lemma 2.2"},{"comment":"In the sentence 'take any w ∈ H^1_0(0,1), v > 0, and define ψµ = ϕ + µw', the symbol v is undefined; presumably it should read 'w > 0' or 'take any w ∈ H^1_0(0,1) with w > 0'.","section":"Section 3, Step 3"},{"comment":"The references to 'Step 3' and 'Step 4' of Proposition 3.1 are off by one: the analogous arguments are Proposition 3.1's Steps 4 and 5, respectively.","section":"Section 4, Steps 4 and 5"},{"comment":"The statement that u(x) > ∥ϕ∥L∞ on the line R × {1/2} relies on the fact that ϕ(1/2) = ∥ϕ∥L∞ for ϕ(t) = t(1−t); this identity is not explicitly noted and could be added for clarity.","section":"Proof of Theorem 1.3"}],"recommendation":"minor_revision","confidential_remarks":"The mathematical content is sound and the central claims are verified. The issues are local presentation gaps and typographical errors. I would be happy to see the paper published after minor revision; no further mathematical review seems necessary."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Here's my read on arXiv:2507.11837. It fills the last missing case (c) in the least-total-curvature classification of steady Euler flows in a strip, and it also produces a stable monotone solution to a semilinear elliptic equation in the same strip with a non-convex superlevel set. Both results are new, and the proof strategy is coherent: build a heteroclinic connection between two ordered 1D minimizers, then construct nonlinearities for which such minimizers exist.\n\nThe variational argument in Section 2 is solid. The truncation to keep approximations between the two 1D minimizers, the reference-point normalization, and the Hamiltonian identity to force convergence to the correct limits are all standard and correct. The new technical content is in Propositions 3.1 and 4.1, where the nonlinearity is engineered so that a second minimizer appears exactly at a threshold lambda*. The total ordering of the minimizer set and the construction of a minimal minimizer above the trivial one are delicate and persuasive. The level-set contradiction in Theorem 1.3 is clean.\n\nThe soft spot is the heavy reliance on earlier work [16]. Proposition 2.1 is explicitly a generalization of [16, Theorem 1.1], and several steps in Propositions 3.1 and 4.1 are only sketched, with references to the previous section or to [16]. Step 5 of Proposition 4.1, the key for the two-positive-minimizer case, is essentially omitted. A referee should ask for the written proof there, though the missing arguments are routine and the stress-test confirms they are fixable. There are also a few typos (the sign of dx1w in Lemma 2.2, 'v > 0' where 'w > 0' is meant in Step 3 of Prop 3.1) and some mislabeled internal step references; none affects the mathematics.\n\nI am satisfied the main claims hold up. The paper completes a classification from [23] and gives a counterexample to quasiconcavity-from-stability in an unbounded convex domain, without touching the bounded-domain conjecture. It deserves a serious referee. I would send it to review, and I'd be comfortable citing it for the existence results once the referee check is done.","headline":"Solid completion of the least-total-curvature classification and a neat counterexample to quasiconcavity-from-stability in a strip; send to review.","tokens_in":17417,"tokens_out":9001,"would_cite":true,"duration_ms":94898,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["35Q31","35Q35","35J61","76B03"],"pacs":[],"model":"deepseek-v4-flash","headline":"The paper constructs the missing case of least total curvature steady Euler flows in a strip and uses the same method to give a stable semilinear solution with non-convex superlevel sets.","keywords":["Steady Euler flow","semilinear elliptic equations","least total curvature flows","heteroclinic solutions","nonconvex superlevel sets","strip domain","stable solutions"],"falsifier":"Numerically solve the one-dimensional minimization $I_\\lambda$ on $H_{0,c}(0,1)$ for the explicit family $F_\\lambda$ at $\\lambda=\\lambda^*$; if a third global minimizer strictly between $\\phi$ and $\\varphi$ is found for either $c=1$ or $c=0$, Assumption 2.1 fails and the heteroclinic construction collapses. Alternatively, directly test the boundary sign pattern of the flow from Theorem 1.2: if $v_1(\\cdot,1)$ does not change sign exactly once (positive for $x_1>0$, negative for $x_1<0$), the flow is not a case (c) least total curvature flow.","tokens_in":16403,"feed_emoji":"🌊","tokens_out":10882,"duration_ms":120027,"temperature":0.7,"pith_summary":"This paper proves existence of the last missing type of least total curvature steady Euler flow in the strip $\\mathbb{R}\\times(0,1)$. In the classification of bounded steady flows by streamline direction, such flows are the non-shear extremal cases of a sharp total-curvature bound; the two previously known types have boundary velocity signs that both change or neither change, and this paper constructs the intermediate case where exactly one boundary changes sign. The proof builds a strictly $x_1$-monotone solution of the semilinear equation $-\\Delta u=f(u)$ with $u=0$ and $u=c$ on the two sides of the strip, connecting two one-dimensional energy minimizers, and then designs a nonlinearity $f$ for which exactly two such ordered minimizers exist. Using the same machinery with $c=0$, the paper constructs a positive, monotone (hence stable) solution with a non-convex superlevel set, which contradicts the generalized quasiconcavity-from-semi-stability conjecture in unbounded convex domains.","feed_headline":"Missing least-total-curvature flow in strip built","feed_subtitle":"The same heteroclinic method yields a stable solution with nonconvex superlevel sets.","key_machinery":"The engine is the variational construction of ordered heteroclinic connections. Starting from two ordered global minimizers of the 1D energy (Assumption 2.1), the authors minimize the full 2D action on truncated strips with Dirichlet data $\\phi,\\varphi$, truncate minimizers to keep the solution between $\\phi$ and $\\varphi$, and pass to the limit with a fixed reference point and the Hamiltonian identity to force the limits to be exactly the two minimizers. The essential design step is the one-parameter family $I_\\lambda$ built on $\\chi(s)^3-\\lambda\\chi(s)^4$; at the threshold $\\lambda^*$ a second global minimizer appears above the trivial one, and a minimal-minimizer selection argument shows the minimizer set is totally ordered, so Assumption 2.1 holds. For the $c=0$ case the two minimizers are both strictly positive and symmetric, which makes the level-set contradiction work: a convex superlevel set would have to be a full horizontal strip $\\mathbb{R}\\times(a,b)$, whose asymptotic limits would force $\\phi(a)=\\varphi(a)$ and $\\phi(b)=\\varphi(b)$, impossible because $\\phi<\\varphi$.","core_discovery":"The central discovery is a variational existence theorem for monotone heteroclinic solutions in the strip: if the one-dimensional energy $I(\\psi)=\\int_0^1(\\tfrac12|\\psi'|^2-F(\\psi))dx_2$ has two global minimizers $\\phi<\\varphi$ with no minimizer of $I$ lying strictly between them, then the boundary-value problem (5) has a solution $u\\in C^{2,\\alpha}$ with $\\partial_{x_1}u>0$, converging to $\\phi$ as $x_1\\to-\\infty$ and to $\\varphi$ as $x_1\\to+\\infty$. Applied to a family of nonlinearities $F_\\lambda(s)=\\chi(s)^3-\\lambda\\chi(s)^4$, this yields the first least total curvature flow in case (c): a bounded smooth steady Euler flow in the strip with $v_2>0$, $v_1(\\cdot,0)\\le v_{1,-}<0$, $v_1(\\cdot,1)>0$ on $(0,\\infty)$ and $v_1(\\cdot,1)<0$ on $(-\\infty,0)$. With $c=0$ and a different nonlinearity, the same construction yields a positive, $x_1$-monotone solution of $-\\Delta u=f(u)$ whose superlevel set $\\{u>\\alpha\\}$ is non-convex for some $\\alpha>0$; since monotone positive solutions are stable, this is a negative answer to the generalized quasiconcavity-from-semi-stability question of [27] in the unbounded convex strip.","pith_inferences":["Editor's inference: the same two-minimizer construction should work for any symmetric double-well potential with an ordered pair of global minimizers, so monotone heteroclinic solutions in strips are likely generic rather than tied to the specific $F_\\lambda$ family.","Editor's inference: the counterexample leaves open whether bounded convex domains can host stable solutions with nonconvex superlevel sets; the strip's unbounded direction may be essential, since the convexity contradiction uses the translation invariance of the domain.","Editor's inference: a direct computation of the total curvature integral (4) for the new case-(c) flow would test whether it is an exact extremal of the sharp lower bound, as in cases (a) and (b), and would make the 'least total curvature' designation quantitative."],"forward_implications":["The classification of least total curvature steady flows in the strip is complete: cases (a), (b), and (c) are each realized by a $C^\\infty$ bounded flow.","Any nonlinearity satisfying Assumption 2.1 yields a strictly $x_1$-monotone heteroclinic solution with prescribed limits $\\phi$ and $\\varphi$, so the construction is a general existence principle rather than a one-off example.","The semilinear solution of Theorem 1.3 is positive, monotone, and therefore stable, so semi-stability of positive solutions in unbounded convex domains does not force convexity of superlevel sets.","The proof shows that $\\{u>\\alpha\\}$ is non-convex for every $\\alpha\\in(0,\\|\\phi\\|_{L^\\infty}]$, so the constructed solution has a whole continuum of non-convex superlevel sets, not an isolated one.","The case-(c) Euler flow fixes the missing streamline diagram in which one boundary layer changes sign exactly once while the other remains of one sign."],"supporting_citations":[{"why":"Supplies the classification theorem (Theorem 1.1) and the total-curvature formulas (3)-(4) that identify case (c) as the missing least total curvature flow.","marker":"[23]"},{"why":"Provides the scheme for constructing monotone heteroclinic solutions in cylinders from two energy minimizers; Proposition 2.1 generalizes [16, Theorem 1.1].","marker":"[16]"},{"why":"Raises the generalized quasiconcavity-from-semi-stability problem that Theorem 1.3 answers negatively in the unbounded strip.","marker":"[27]"},{"why":"Supplies the principal eigenvalue and maximum principle tools used to prove strict $x_1$-monotonicity of the heteroclinic solution.","marker":"[3]"},{"why":"Provides the Hamiltonian identity used to identify the limits of the heteroclinic solution as global minimizers of the 1D energy.","marker":"[22]"}],"fun_headline_variants":["Missing Euler flows found via heteroclinic method","Least-curvature flows in strip constructed","Monotone solutions yield nonconvex level sets","Variational trick builds stable flows and nonconvex sets","Heteroclinic minimization solves Euler strip case"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is Assumption 2.1: the one-dimensional energy $I$ has two global minimizers, one strictly below the other, with no minimizer between them; the paper's verification of this premise is the technical heart, and parts of it (notably the minimal-minimizer selection in Proposition 4.1) are only sketched or left to analogy with [16], so a gap in those steps would undermine both theorems.","fun_headline_variants_meta":{"raw":{"variants":["Missing Euler flows found via heteroclinic method","Least-curvature flows in strip constructed","Monotone solutions yield nonconvex level sets","Variational trick builds stable flows and nonconvex sets","Heteroclinic minimization solves Euler strip case"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000608,"raw_usage":{"total_tokens":2848,"prompt_tokens":975,"completion_tokens":1873,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":591,"completion_tokens_details":{"reasoning_tokens":1798}},"tokens_in":591,"tokens_out":1873,"duration_ms":15373,"temperature":1.0,"reasoning_tokens":1798,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T17:03:20.302002+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Numerically solve the one-dimensional minimization $I_\\lambda$ on $H_{0,c}(0,1)$ for the explicit family $F_\\lambda$ at $\\lambda=\\lambda^*$; if a third global minimizer strictly between $\\phi$ and $\\varphi$ is found for either $c=1$ or $c=0$, Assumption 2.1 fails and the heteroclinic construction collapses. Alternatively, directly test the boundary sign pattern of the flow from Theorem 1.2: if $v_1(\\cdot,1)$ does not change sign exactly once (positive for $x_1>0$, negative for $x_1<0$), the flow is not a case (c) least total curvature flow.","supporting_citations":[{"cited_title":"De Regibus and D","cited_arxiv_id":null,"evidence_quote":"Provides the scheme for constructing monotone heteroclinic solutions in cylinders from two energy minimizers; Proposition 2.1 generalizes [16, Theorem 1.1]."},{"cited_title":"Hamel, N","cited_arxiv_id":null,"evidence_quote":"Raises the generalized quasiconcavity-from-semi-stability problem that Theorem 1.3 answers negatively in the unbounded strip."},{"cited_title":"Berestycki, L","cited_arxiv_id":null,"evidence_quote":"Supplies the principal eigenvalue and maximum principle tools used to prove strict $x_1$-monotonicity of the heteroclinic solution."},{"cited_title":"Gui, Hamiltonian identities for elliptic partial differential equations, J","cited_arxiv_id":null,"evidence_quote":"Provides the Hamiltonian identity used to identify the limits of the heteroclinic solution as global minimizers of the 1D energy."}],"review_version":1}