{"id":"d341db0d-aa53-477e-bf2e-284af11f1abf","arxiv_id":"2506.22371","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Energy local minimizers with prescribed L2 norm exist in the mass-supercritical NLS on R^N x M^k; for small mass they are the Euclidean ground states, and in certain cases they become nontrivial along M^k.","lead":"This paper proves that for a nonlinear Schrödinger equation on the product of Euclidean space and a compact manifold, local minimizers of the energy exist at small mass even where global minimizers do not. It also proves when these minimizers are constant along the compact manifold and gives examples where they are not.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The proof that rho*_tr>0 (Theorem 1.2(2)) delegates the key rigidity step to [16, Lemma 3.6] without verifying its hypotheses for the lambda-dependent constrained minimizers of E_lambda; the dichotomy in Theorem 1.2(2)-(3) depends on this step.","rationale":"The central theorem is an existence and dichotomy result for local minimizers, and the architecture is convincing: Section 3 constructs minimizers through a mountain-pass type argument, Lemma 4.1 gives the structural monotonicity of the thresholds, and Section 5 provides a plausible instability mechanism for Z_rho. The most load-bearing point is the proof that rho*_tr > 0, namely that all sufficiently small local minimizers are y-independent. The paper tries to prove this by showing that any sequence of rescaled minimizers converges to Z1 and then invoking [16, Lemma 3.6]. That final invocation is the place where the argument is least secure: the cited lemma was proved in a mass-subcritical, globally coercive setting, while the present admissible set carries a lambda-dependent bound, and the local uniqueness needed to turn convergence into exact rigidity is not stated or proved. This is not an accusation of error; the result is likely true and the missing details may be standard. But because Theorem 1.2(2) is precisely the assertion that the small-mass local minimizers coincide with the Euclidean family, and Theorem 1.2(3) only becomes meaningful once rho*_tr > 0, this gap is the right point to demand an explicit check. The undocumented CAS verification in Proposition 1.14 is a legitimate reproducibility concern, but it is secondary: Proposition 1.12 already provides broad examples of nontrivial minimizers, and a failure of the k=4,5 check would not destroy Theorem 1.2. The reader's CONDITIONAL verdict already reflects this incompleteness, so no verdict change is needed.","tokens_in":18656,"tokens_out":25839,"duration_ms":279541,"concrete_test":"Derive the full statement of [16, Lemma 3.6] and prove the following local rigidity lemma for the rescaled problem: for every alpha in (1.10) there exist Lambda >= 1 and delta > 0 such that, for every lambda >= Lambda and every critical point u in S1 of E_lambda on S1 satisfying the admissible bound in (4.4), ||u - Z1 o tau||_{H1(R^N x M)} < delta for some Euclidean translation tau, and E_lambda(u) <= I1 + 1/Lambda, one has grad_y u = 0. The lemma must be proved using only Sobolev subcriticality and (4.7)-(4.8); if any step needs global coercivity of E_lambda or alpha < 4/(N+k), that is a missing hypothesis. Equivalently, run a Lyapunov-Schmidt reduction near the orbit of Z1 to exclude bifurcating y-dependent branches for all large lambda; the existence of such a branch for arbitrarily large lambda would disprove Theorem 1.2(2).","verdict_should_be":"UNCHANGED","load_bearing_attack":"Section 4, after Lemma 4.4, asserts that u_lambda_j can be completed 'as in [16, Section 3]' to get strong convergence to Z1 and then grad_y u_lambda_j = 0 for all large j, citing [16, Lemma 3.6]. This is the step that makes Theorem 1.2(2) meaningful: it converts convergence of one subsequence into the statement that every sufficiently small local minimizer is y-independent. Two hypotheses are needed and neither is checked. First, [16, Lemma 3.6] was proved for global minimizers in the mass-subcritical regime; here the admissible set (4.4) has the lambda-dependent moving bound (1/lambda)||grad_x u||^2 + ||grad_y u||^2 < t*, so coercivity and compactness do not transfer automatically. The bounds (4.7)-(4.8) do not by themselves give a uniform H1-neighbourhood of the orbit of Z1 adapted to this constraint. Second, to conclude that a critical point at energy level I1 near Z1 is exactly Z1 (up to translation and phase), one needs a local uniqueness or nondegeneracy statement for the Euclidean ground state; Section 2 only quotes uniqueness of the positive solution, and the paper says the only assumption in [16, Lemma 3.6] is Sobolev subcriticality. If that lemma relies on more structure, or if the moving bound invalidates its proof, then rho*_tr>0 is not established and the threshold dichotomy in Theorem 1.2(2)-(3) does not follow from the displayed estimates.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper studies the mass-supercritical but Sobolev-subcritical NLS energy E(u)=1/2||∇u||_2^2 - 1/(2+α)||u||_{2+α}^{2+α} on R^N × M^k, with prescribed L^2 mass ρ, restricted to the sublevel set ||∇u||_2^2 < t_* ρ^2. The main result, Theorem 1.2, asserts existence of local minimizers for small ρ (part 1), their y-independence and coincidence with the Euclidean ground states for ρ < ρ*_tr (part 2), and nontrivial y-dependence with m_ρ < I_ρ for intermediate ρ when ρ*_tr < ρ*_ex (part 3). The proof uses Gagliardo-Nirenberg inequalities, a mountain-pass type geometry for existence, a scaling λ = ρ^{-4α/(4-αN)} that penalizes ∇_y, and a threshold dichotomy. Propositions 1.12 and 1.14 provide sufficient conditions for ρ*_tr < ρ*_ex via small first eigenvalue of -Δ_y and explicit computations on spheres.","tokens_in":18959,"tokens_out":37254,"duration_ms":341425,"significance":"The paper addresses an open question from [16] and gives a natural local-minimizer analogue of the mass-subcritical ground-state threshold. The explicit formulas for ρ*_ex, the upper bound for ρ*_tr, the metric-deformation result, and the sphere examples are concrete and useful, and the central existence mechanism and scaling identities appear sound. However, the proof of the key rigidity step ρ*_tr > 0 is delegated to [16, Lemma 3.6] without a statement or verification of its hypotheses, and the beta-function computation in Section 7 contains an algebraic error. Both points need to be fixed before the results can be considered established.","major_comments":[{"comment":"The step that converts Lemma 4.4 into the conclusion ρ*_tr > 0 is not proved in the manuscript. After showing m_λj → I_1 and λ_j ||∇_y u_λj||^2 → 0, the text asserts that u_λj → Z_1 strongly and then invokes [16, Lemma 3.6] to conclude ∇_y u_λj = 0 for large j, saying that the only assumption is Sobolev subcriticality. This is load-bearing for Theorem 1.2(2)-(3), but two hypotheses are not verified. First, the admissible set (4.4) carries the λ-dependent moving bound λ^{-1}||∇_x u||^2 + ||∇_y u||^2 < t_*, whereas [16, Lemma 3.6] is stated for global minimizers of the unconstrained problem in the mass-subcritical regime; the estimates (4.7)-(4.8) do not by themselves provide a uniform H^1-neighbourhood of the orbit of Z_1 that is compatible with this moving constraint. Second, the identification of u_λj with Z_1 up to symmetries requires a local uniqueness or nondegeneracy statement for the Euclidean ground state of (2.3); Section 2 only records uniqueness of the positive solution, not nondegeneracy. The sentence 'the main assumption is Sobolev subcritical' does not replace a statement of the lemma and a check of its hypotheses. Until this step is supplied, ρ*_tr > 0 is not established.","section":"Section 4, after Lemma 4.4 (Eqs. (4.7)-(4.8))"},{"comment":"The displayed identity (2/α) B(1/2,2/α) B(1/2,(k+1)/2) = [Γ(2/α+1)/Γ(2/α+1/2)] [Γ((k+1)/2+1/2)/Γ((k+1)/2)] is not correct with the Beta function definition given in that section. The left-hand side equals π [Γ(2/α+1)/Γ(2/α+1/2)] [Γ((k+1)/2)/Γ(k/2+1)], so the subsequent bound T_1 ≤ ((k+1)/2)^{2/(k-1)} does not follow from the displayed computation. Since the proof of Proposition 1.14 for k ≥ 6 relies on this bound, the computation must be corrected or replaced; if the bound remains true it needs a different argument.","section":"Section 7, displayed identity before (7.3)"}],"minor_comments":[{"comment":"The second variation used is the unconstrained one; for a path in S_ρ one obtains E''(Z_ρ)[h,h] - ωρ^2 because h is tangent and the projection contributes -ωρ^2. The condition E'' < 0 is still sufficient since ω > 0 for Z_ρ, but this should be stated explicitly.","section":"Section 5, Proposition 5.1"},{"comment":"The verification for k = 4,5 via Computer Algebra System is not reproducible; please provide the script or list the inequalities checked.","section":"Section 7, proof of Proposition 1.14"},{"comment":"The text says '∇ u_λj ≡ 0 for j sufficiently large', but the preceding argument only yields convergence along a subsequence; clarify that a subsequence argument suffices for the contradiction.","section":"Section 4, after Lemma 4.4"},{"comment":"In the mass-critical case α = 4/(N+k), t* is chosen 'suitably large' without specifying the relation between t* and ε; please spell out the choice.","section":"Section 3, Lemma 3.2"},{"comment":"The embedding statement '... yields, for any k ≥ 1' after writing k ≥ 3 mixes cases; treat k = 1,2 separately.","section":"Lemma 4.4"}],"recommendation":"major_revision","confidential_remarks":"The main risk is the unverified use of [16, Lemma 3.6]; the authors should state the lemma and prove the required local uniqueness in the λ-dependent constrained setting. The beta-function identity in Section 7 appears to be a substantial typo; check whether the final inequalities survive with the correct factor. If these points are repaired, the paper should be acceptable."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Let me give you the short version: this is a good paper that does what it says. It answers the open question from [16] by showing that in the mass-critical and supercritical (but Sobolev subcritical) range, the energy on the mass sphere has local minimizers for small L2-mass, and that these are either y-independent Euclidean ground states or genuinely y-dependent, depending on a threshold. The main theorem is plausible and the proof architecture is sound.\n\nWhat's actually new: the constrained gradient ball (1.9) that allows a local minimization problem where the global problem is unbounded below; a mountain-pass lemma (Lemma 3.2) and a uniform-in-lambda estimate (Lemma 4.4) with a refined Gagliardo-Nirenberg inequality; the threshold dichotomy in Theorem 1.2. The mass-critical case is handled by a limiting argument. These are real contributions.\n\nWhere I'd push back: the proof of rho*_tr > 0, the step that turns convergence of one subsequence into a statement about all small minimizers, is delegated to [16, Section 3] and specifically [16, Lemma 3.6]. The authors claim the only needed assumption is Sobolev subcriticality, and they do supply the modified vanishing argument in (3.7)-(3.8). But they don't verify that the cited lemma's hypotheses hold for the constrained functionals E_lambda with the lambda-dependent moving bound (4.4). This is not a fatal flaw—nothing suggests the lemma fails—but it's a gap in self-containedness that a referee should ask them to close, either by reproducing the argument or by stating the lemma explicitly and checking its hypotheses.\n\nThe second soft spot is Proposition 1.14. The cases k=4,5, away from the lower endpoint, are checked with a Computer Algebra System, but no script, output, or reproducible details are provided. In Remark 1.15 they mention an analytic proof for part of the range, but the remainder is undocumented. That should be fixed before publication.\n\nAlso, the paper leans on [16] for compactness of minimizing sequences. That's acceptable, but they should make the precise statements clear. The reader's weakest-assumption note is a fair point.\n\nOverall: this paper deserves a serious referee. The central result is important for the normalized NLS community, and the authors are clearly in control of the material. I'd recommend sending it to peer review and asking for the two gaps above to be addressed; then it should be fine.\n\nIf you want a quick take: I'd bring this to our reading group if anyone works on PDEs or variational methods. I'd cite it if I were working on normalized solutions. My verdict is conditional, leaning positive.","headline":"Solid, important paper on local minimizers for mass-supercritical NLS on product spaces; the proof is convincing but has two documentation gaps—the rigidity step via [16] and a CAS check—that a referee should ask to close.","tokens_in":19556,"tokens_out":4488,"would_cite":true,"duration_ms":43561,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["58J05","35Q55","58J70"],"pacs":[],"model":"deepseek-v4-flash","headline":"The paper establishes a two-threshold picture for the mass-supercritical NLS energy on $\\mathbb{R}^N \\times M^k$: local minimizers of prescribed mass exist, are flat along $M^k$ for very small mass, and become $y$-dependent once the mass…","keywords":["Normalized solutions","elliptic PDEs on manifolds","constrained critical points","waveguide manifolds","mass-supercritical NLS","local minimizers","Gagliardo-Nirenberg inequality"],"falsifier":"On $\\mathbb{R} \\times \\mathbb{S}^3$ with the round metric and an exponent just below $2$, the paper's condition (7.3) fails; running its computer-algebra check at higher precision, or numerically minimizing the rescaled functional $E_\\lambda$ for a sequence $\\lambda_j \\to \\infty$ and observing a limit with nonzero $\\nabla_y$ component, would settle whether $\\rho^*_{tr}>0$ holds in that case.","tokens_in":18407,"feed_emoji":"🌀","tokens_out":14584,"duration_ms":132266,"temperature":0.7,"pith_summary":"This paper proves that the mass-supercritical Schrödinger energy on a product space $\\mathbb{R}^N \\times M^k$, where $(M^k,g)$ is a compact Riemannian manifold, admits local minimizers of prescribed $L^2$-norm even though the energy is unbounded below on the whole mass sphere. The regime treated is $4/(N+k) \\le \\alpha < \\min\\{4/N,\\, 4/(N+k-2)\\}$, complementing an earlier mass-subcritical global-minimization result. The authors show there are two thresholds: below a triviality threshold every local minimizer is constant along $M^k$ and coincides with the Euclidean ground state, while above it, when the thresholds separate, every local minimizer has nontrivial dependence on the compact direction and the constrained infimum lies strictly below the Euclidean one. This provides a variational characterization of the family expected to be stable in the mass-supercritical range, where no global minimizer exists.","feed_headline":"Mass-supercritical NLS gains local minimizers on product spaces","feed_subtitle":"Small mass forces flat waves along the compact direction; past a threshold, waves bend and lower the energy.","key_machinery":"The engine is a rescaling that fixes the $L^2$-norm: writing $u = \\rho^{4/(4-\\alpha N)} v(\\rho^{2\\alpha/(4-\\alpha N)} x, y)$, the energy becomes, up to a positive factor, $E_\\lambda(v) = \\int_{M^k}\\int_{\\mathbb{R}^N} \\left(\\tfrac12|\\nabla_x v|^2 + \\tfrac{\\lambda}{2}|\\nabla_y v|^2 - \\tfrac{1}{2+\\alpha}|v|^{2+\\alpha}\\right)$, with $\\lambda = \\rho^{-4\\alpha/(4-\\alpha N)}$, and the gradient constraint becomes $\\tfrac1\\lambda\\|\\nabla_x v\\|_2^2 + \\|\\nabla_y v\\|_2^2 < t_*$. Thus larger $\\lambda$, corresponding to smaller mass, penalizes dependence on the compact factor. The paper pairs this rescaling with the Gagliardo-Nirenberg inequality on $\\mathbb{R}^N \\times M^k$, with $\\theta(\\alpha) = (N+k)\\alpha/2$, to locate the existence threshold $\\rho^*_{ex}$, and with the first nonzero eigenvalue $\\mu_1$ of $-\\Delta_y$ on $M^k$ to bound $\\rho^*_{tr}$ from above through the second variation of $E$ at the Euclidean ground state $Z_\\rho$ along the direction $\\varphi_1(y)Z_\\rho(x)$.","core_discovery":"The central claim is that the constrained local problem $m_\\rho = \\inf\\{E(u): u \\in S_\\rho,\\, \\|\\nabla_{x,y}u\\|_2^2 < t_* \\rho^2\\}$ is well posed in the mass-supercritical range $4/(N+k) \\le \\alpha < \\min\\{4/N,\\, 4/(N+k-2)\\}$. Theorem 1.2 asserts that for $0<\\rho<\\rho^*_{ex}$ the infimum is achieved by a solution of the normalized equation $-\\Delta_x u - \\Delta_y u + \\omega u = u|u|^\\alpha$ with $\\|u\\|_2 = \\rho$, that for $0<\\rho<\\rho^*_{tr}$ every minimizer satisfies $\\nabla_y u \\equiv 0$ and equals, up to symmetries, the Euclidean ground state, so that $m_\\rho = I_\\rho$, and that if $\\rho^*_{tr} < \\rho^*_{ex}$ then for intermediate masses every minimizer is nontrivial in $y$ and $m_\\rho < I_\\rho$. The paper also proves that the nontrivial regime genuinely occurs: for any compact manifold of dimension $k\\ge 2$ one can choose a metric making the thresholds differ, and on $\\mathbb{R} \\times \\mathbb{S}^k$ with the round metric this happens for every admissible exponent when $k\\ge 4$ and in a neighbourhood of the mass-critical endpoint when $k=3$.","pith_inferences":["The unproved transfer of the convergence-to-$Z_1$ lemma is the main hidden step; if it fails for some $\\alpha$, the small-mass flatness could fail even where existence holds, and the paper's own negative computer-algebra check for $k=3$, $\\alpha$ large on $\\mathbb{S}^3$ marks the spot to test.","The same constrained-infimum scheme should extend to the Sobolev-critical endpoint $\\alpha = 4/(N+k-2)$, where the paper expects local minimizers below a threshold; a numerical search for admissible competitors there would show whether the mountain-pass structure survives.","The threshold $\\rho^*_{tr}$ is likely a bifurcation point at which flat and $y$-dependent minimizers exchange stability; numerical continuation in $\\rho$ for a fixed manifold and exponent would map the actual bifurcation diagram and test the coexistence behaviour cited in the paper's Remark 1.6.","Since the criterion (5.4) is only sufficient, using higher eigenvalues $\\mu_k$ or anisotropic Gagliardo-Nirenberg bounds should widen the window $(\\rho^*_{tr}, \\rho^*_{ex})$, as the paper itself hints in Remark 3.6."],"forward_implications":["For every $0<\\rho<\\rho^*_{ex}$ the normalized equation has a solution realizing $m_\\rho$, giving a variational description of the mass-supercritical ground-state family even though $E$ is unbounded below on $S_\\rho$.","For $0<\\rho<\\rho^*_{tr}$ every local minimizer is constant along $M^k$ and coincides with the Euclidean ground state up to symmetries, so the compact manifold is invisible to the minimizers at small mass.","If the thresholds separate, then in the intermediate interval every minimizer has $\\nabla_y u \\not\\equiv 0$ and $m_\\rho < I_\\rho$; the presence of the compact factor strictly lowers the minimal energy.","For any compact manifold of dimension $k\\ge 2$ there is a choice of metric making the nontrivial regime occur, and on $\\mathbb{R} \\times \\mathbb{S}^k$ with the round metric it occurs for all admissible exponents when $k\\ge 4$ and near the mass-critical endpoint when $k=3$.","The local minimizers are the natural candidates for conditionally orbitally stable solitary waves of the evolution equation, as the paper notes following the stability analysis of the mass-subcritical case."],"supporting_citations":[{"why":"Supplies the mass-subcritical ground-state theory on product spaces and the compactness and convergence lemmas, including the step identifying rescaled minimizers with $Z_1$, that the proof adapts.","marker":"[16]"},{"why":"Provides the Sobolev and Gagliardo-Nirenberg inequalities on manifolds, Theorems 7.1 and 7.7, whose constants $A,B$ define the existence threshold and the sphere estimates.","marker":"[7]"},{"why":"Gives the Euclidean ground states $Z_\\rho$, their uniqueness up to symmetries, and the scaling identities (2.6)--(2.7) used throughout the paper.","marker":"[10]"},{"why":"Establishes the local-minimizer approach for $L^2$-critical and supercritical NLS on bounded domains, the paradigm extended here to product spaces.","marker":"[11]"},{"why":"Shows that any compact manifold of dimension $k\\ge 2$ admits metrics with arbitrarily small first nonzero Laplace eigenvalue, driving Proposition 1.12.","marker":"[4]"},{"why":"Provides existence and stability of standing waves for supercritical NLS with partial confinement, a related setting motivating local minimizers in absence of global ones.","marker":"[1]"}],"fun_headline_variants":["Supercritical NLS minimizers bend on compact spaces at large mass","Product-space NLS: small mass flat, bigger mass twists","Mass-supercritical NLS on product spaces: local minimizers appear","NLS local minimizers on product spaces: flat for small mass, curved for large","When do NLS minimizers on product spaces become nonconstant in compact directions?"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The flatness result for very small mass depends on transferring a convergence lemma from the global-minimizer setting to the constrained local problem with moving gradient bound, and on the uniqueness up to translations and phase shifts of the Euclidean ground state for every exponent in the range; the paper does not prove the needed nondegeneracy or the transfer.","fun_headline_variants_meta":{"raw":{"variants":["Supercritical NLS minimizers bend on compact spaces at large mass","Product-space NLS: small mass flat, bigger mass twists","Mass-supercritical NLS on product spaces: local minimizers appear","NLS local minimizers on product spaces: flat for small mass, curved for large","When do NLS minimizers on product spaces become nonconstant in compact directions?"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000204,"raw_usage":{"total_tokens":1467,"prompt_tokens":1104,"completion_tokens":363,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":720,"completion_tokens_details":{"reasoning_tokens":266}},"tokens_in":720,"tokens_out":363,"duration_ms":4480,"temperature":1.0,"reasoning_tokens":266,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T22:07:26.229530+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"On $\\mathbb{R} \\times \\mathbb{S}^3$ with the round metric and an exponent just below $2$, the paper's condition (7.3) fails; running its computer-algebra check at higher precision, or numerically minimizing the rescaled functional $E_\\lambda$ for a sequence $\\lambda_j \\to \\infty$ and observing a limit with nonzero $\\nabla_y$ component, would settle whether $\\rho^*_{tr}>0$ holds in that case.","supporting_citations":[{"cited_title":"Terracini, N","cited_arxiv_id":null,"evidence_quote":"Supplies the mass-subcritical ground-state theory on product spaces and the compactness and convergence lemmas, including the step identifying rescaled minimizers with $Z_1$, that the proof adapts."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the Sobolev and Gagliardo-Nirenberg inequalities on manifolds, Theorems 7.1 and 7.7, whose constants $A,B$ define the existence threshold and the sphere estimates."},{"cited_title":"Molle, G","cited_arxiv_id":null,"evidence_quote":"Gives the Euclidean ground states $Z_\\rho$, their uniqueness up to symmetries, and the scaling identities (2.6)--(2.7) used throughout the paper."},{"cited_title":"Noris, H","cited_arxiv_id":null,"evidence_quote":"Establishes the local-minimizer approach for $L^2$-critical and supercritical NLS on bounded domains, the paradigm extended here to product spaces."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Shows that any compact manifold of dimension $k\\ge 2$ admits metrics with arbitrarily small first nonzero Laplace eigenvalue, driving Proposition 1.12."},{"cited_title":"Bellazzini, N","cited_arxiv_id":null,"evidence_quote":"Provides existence and stability of standing waves for supercritical NLS with partial confinement, a related setting motivating local minimizers in absence of global ones."}],"review_version":1}