{"id":"9b0e0724-9e6a-4e5d-8c90-3aea1241a38c","arxiv_id":"2411.14340","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":8.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"A new condition called quasi-parallel mean curvature yields a unique canonical foliation by spheres of any metric close to the product metric on R^k x S^(n-k), giving a normal form for bubblesheets.","lead":"This paper defines a new curvature condition, quasi-parallel mean curvature (QPMC), and proves that on R^k x S^(n-k) with any sufficiently close metric there is a unique foliation by almost-round spheres satisfying it. The foliation gives a canonical normal form for bubblesheets, the high-codimension cylinder-like regions expected to dominate singularities in mean curvature flow and Ricci flow.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Theorem 1.6's gluing argument does not prove that a δ-vertical QPMC leaf from one chart lies inside the neighboring chart, so the invoked uniqueness in Proposition 5.5 is inapplicable as written.","rationale":"The central result Theorem 1.2 is proved by a clean implicit-function-theorem argument whose linearization is the scalar Laplacian; I found no substantive flaw there. The variation formulae in Section 3 are consistent (I checked the spectral perturbation computation in Proposition 3.2), and the QPMC bootstrap in Proposition 4.7 is standard. The decisive weakness is the passage from local to global in Theorem 1.6: the gluing step needs a geometric containment statement that is asserted rather than proved. This is load-bearing because Theorem 1.6 is what produces the canonical foliation of bubblesheet regions; without it the abstract's advertised application is not established. The gap is fixable and does not affect Theorem 1.2, so conditional acceptance remains the right call. My concrete test (a direct containment lemma or a rewrite of the gluing via Lemma 5.4) would settle it. I agree with the reader's identification of this as the weakest assumption.","tokens_in":19766,"tokens_out":20965,"duration_ms":184603,"concrete_test":"Prove or disprove a containment lemma: for L ≥ 1000, if F: B^k(0,L)×S^{n-k}→M is an (ε,L,n-k)-chart and Σ is a δ-vertical QPMC sphere intersecting F(B^k(0,90)×S^{n-k}), then F^{-1}(Σ) ⊂ B^k(0,L-20)×S^{n-k} for ε,δ sufficiently small. A proof would use Lemma 5.1 to write Σ as a graph u: S^{n-k}→R^k with ||du||_{C^3} ≤ C(ε+δ), then bound |u - u(ω_x)| by diam(S^{n-k})·sup|du| and the center displacement by the diameter bound to conclude the z-coordinates stay within 90+π+Cδ. If the lemma fails, replace the invocation of Proposition 5.5 in Theorem 1.6 by a direct application of Lemma 5.4 after first proving the pulled-back sphere is closed in the chart; verify the resulting gluing argument.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The proof of Theorem 1.6 (Section 5) glues local foliations by asserting that the pullback of a leaf Σ_q(x) by a chart F_p is of class S(δ) and then invoking the uniqueness statement in Proposition 5.5 to conclude Σ_q(x) ∈ F_p^+. Proposition 5.5's uniqueness applies only to spheres Σ contained in the open set B on which the local foliation is defined ('if Σ ⊂ B is of class S_g(δ) and has QPMC, then Σ is a leaf'). The proof knows only x ∈ B_p ∩ B_q; it never establishes that the entire sphere F_p^{-1}(Σ_q(x)) lies in the chart domain B^k(0,L-10)×S^{n-k} (nor even in the smaller B). The same gap appears in the final paragraph: a δ-vertical QPMC sphere intersecting C is declared a leaf without proving it is contained in the chart. This is not a fatal flaw—a containment lemma is plausible from the diameter bound diam(Σ)/¯r ≤ 10π in Definition 1.5, Lemma 5.1's graphical estimates, and L ≥ 1000—but as written the gluing step is unsupported. Since Theorem 1.6 is the mechanism that turns the local Theorem 1.2 into the advertised 'normal form for bubblesheets', this is the most load-bearing gap.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper introduces a new curvature condition for codimension-k submanifolds, called quasi-parallel mean curvature (QPMC), defined by requiring the mean curvature vector H to lie in the spectral subspace of the normal Laplacian corresponding to eigenvalues below lambda_{k+1}. The main local result, Theorem 1.2, shows that on M = R^k x S^{n-k}, any metric sufficiently close in C^{ell-1,gamma} to the product metric g0 admits a unique smooth foliation by embedded (n-k)-spheres with QPMC that are delta-close to the standard slices; the proof uses the implicit function theorem after quotienting out translations, with the linearization computed exactly as the sphere Laplacian. Theorem 1.6 upgrades this to a global statement for (epsilon, L, n-k)-cylindrical regions, asserting the existence of a canonical QPMC foliation covering the region and a uniqueness statement for delta-vertical QPMC spheres meeting it. Section 6 gives examples of metrics on R^2 x S^1, arbitrarily close to g0, whose canonical QPMC foliations do not consist of parallel mean curvature spheres, and Section 7 sketches applications to mean curvature flow normal forms and topology.","tokens_in":19971,"tokens_out":11804,"duration_ms":113739,"significance":"If the proofs are completed, this is a useful and original contribution: QPMC provides a natural intermediate condition between CMC and parallel mean curvature that is adapted to codimension k >= 2, and the foliation theorem gives a canonical normal form for bubblesheet regions in geometric flows. A notable strength is that the construction is parameter-free: the implicit function theorem argument is built around the explicit background metric g0, the linearization is computed exactly as the sphere Laplacian, and all constants depend only on n, ell, gamma, delta. The examples in Section 6 showing non-PMC QPMC foliations are concrete and support the necessity of the new condition. The main gaps are technical rather than conceptual: the gluing argument in Theorem 1.6 needs a containment lemma, and the regularity bootstrap in Proposition 4.7 needs a rigorous treatment of the nonlocal projection Q.","major_comments":[{"comment":"The gluing step applies the uniqueness clause of Proposition 5.5 to the pulled-back leaf F_p^{-1}(Sigma_q(x)), but Proposition 5.5's uniqueness is conditional on the entire sphere being contained in the chart domain B between B^k(0,L-20) x S^{n-k} and B^k(0,L-10) x S^{n-k}. The proof only establishes that F_p^{-1}(x) lies in B^k(0,90) x S^{n-k} for the point x in the overlap; it does not establish that the whole sphere F_p^{-1}(Sigma_q(x)) is contained in B. The same omission occurs in the final paragraph, where a delta-vertical QPMC sphere intersecting C is declared a leaf without first proving that the sphere lies in the local chart. This is load-bearing because this identification of leaves is exactly how the local foliations are glued and how the global uniqueness assertion is obtained. The gap is plausibly closed using the diameter bound diam(Sigma)/bar-r <= 10 pi in Definition 1.5 together with Lemma 5.1 and the choice L >= 1000, but the argument is absent as written.","section":"Section 4, Proposition 4.7"},{"comment":"The proof of the C^infty bootstrap treats the equation (1 - Q)(H) = 0 as though it were a local elliptic system for u, but Q is a nonlocal spectral projection of -Delta^perp on the pulled-back normal bundle of Sigma_u, so its regularity depends on u in a way that the displayed argument does not control. From u in C^{ell,gamma} one obtains eigensections in C^{ell-1,gamma} by Lemma 2.1 and hence Q(H) in C^{ell-1,gamma}; the equation H = Q(H) then yields u in C^{ell+1,gamma} by Schauder theory only if one proves that the coefficients of H as a second-order operator on u are sufficiently regular and that the projection Q remains controlled along the iteration. The current proof is only a sketch and does not supply this induction. Since Theorem 1.2 claims that the foliation is smooth, this regularity statement is load-bearing and needs a detailed proof or a precise citation of a theorem that covers this nonlocal quasilinear setting.","section":"Section 4, Proposition 4.7"}],"minor_comments":[{"comment":"The proof refers to 'Lemma 4.7' where Proposition 4.7 is meant; the bootstrap result is a proposition, not a lemma.","section":"Corollary 4.8"},{"comment":"The displayed assumption 'ell /greaterorequalslant 3' is a rendering artifact and should read 'ell >= 3'.","section":"Proposition 4.6"},{"comment":"The definition of delta-vertical should specify whether diam_g(Sigma) is the intrinsic or the ambient diameter; the gluing argument in Theorem 1.6 depends on a quantitative relation between this diameter and the coordinate variation in a chart.","section":"Definition 1.5"},{"comment":"The text refers to 'Theorem 7.1' when citing the neck detection statement; the cited statement is Proposition 7.1.","section":"Section 6"},{"comment":"In the graphicality proof, the comparison 'up to errors ... the left-hand side is q_ab and the right-hand side is delta_ab' should be replaced by an explicit estimate, since it is the core of the claim that det(q_ab) > 0.","section":"Lemma 5.1"},{"comment":"The sentence 'There is reason to believe that all of the m do, up to isotopy' is unclear and should be rewritten, for example as 'all such m-dimensional submanifolds arise, up to isotopy', if that is the intended claim.","section":"Section 7, Proposition 7.2"}],"recommendation":"major_revision","confidential_remarks":"This is a promising paper with a sound core idea and a parameter-free local existence proof. The two gaps identified in the report are repairable: a containment lemma in Section 5 and a rigorous regularity bootstrap in Section 4. I recommend major revision rather than rejection; the authors should be asked to supply these arguments explicitly before the paper is accepted."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Read the Lagacé–Lynch paper. The core is real: they define QPMC, prove that metrics close to the product on R^k × S^{n-k} admit a unique foliation by QPMC spheres, and the proof is a clean IFT argument with a correct linearization at the product metric. This is genuinely new—prior foliation results were codimension-one CMC. The examples in Section 6 showing that PMC foliations need not exist are a nice touch.\n\nThe soft spots are at the margins, not the center. The gluing step in Theorem 1.6 has a small gap: they invoke the uniqueness statement in Proposition 5.5 for a leaf pulled back to a chart without proving the leaf lies inside the chart's domain. The diameter bound in the δ-vertical definition and L≥1000 make this true, but they should say it. A referee should ask for that one-line argument.\n\nThe regularity bootstrap in Proposition 4.7 is sketched. They treat the nonlocal projection Q as if the equation were local elliptic, and the gain of one derivative from Schauder is plausible but not fully written out. The dependence of the eigensections on u needs a bit more care. Again, likely fixable, but it should be detailed.\n\nSection 7's Proposition 7.2 is more of a heuristic statement than a proof; the topological conclusion about isotopy needs more support. And the Section 6 construction is sketched rather than fully proved, though the idea is sound.\n\nNo circularity, no fitted parameters. The main theorem is parameter-free. The citations are appropriate; the external results (Schauder, bumpy metrics, Berger sphere stability) are standard and used correctly.\n\nThis is for geometric analysts working on MCF and Ricci flow singularities. The central result is correct in structure and the gaps are patchable. Send it to review, with instructions to ask for the containment argument in Theorem 1.6 and a fuller regularity bootstrap.","headline":"Solid new QPMC foliation theorem with a minor gluing gap that is easily patched.","tokens_in":20556,"tokens_out":3952,"would_cite":true,"duration_ms":35574,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["53C40","53C42","53C44","58J50"],"pacs":[],"model":"deepseek-v4-flash","headline":"A new curvature condition gives near-product cylinder metrics a unique foliation by almost-round spheres.","keywords":["quasi-parallel mean curvature","canonical foliation","bubblesheets","normal Laplacian","spectral projection","high-codimension submanifolds","mean curvature flow","Riemannian geometry"],"falsifier":"In the $\\mathbb{R}^2 \\times \\mathbb{S}^1$ example of the paper, compute the eigenvalues of the normal Laplacian on the QPMC leaves as the metric approaches the product metric; if $\\lambda_{k+1}-\\lambda_k$ approaches zero on a sequence of leaves, the projection $Q$ stops having constant rank $k$ and the frame construction behind the implicit-function-theorem proof breaks down, contradicting the claimed uniform foliation near that limit.","tokens_in":19505,"feed_emoji":"🫧","tokens_out":14213,"duration_ms":117539,"temperature":0.7,"pith_summary":"The paper introduces quasi-parallel mean curvature (QPMC), a spectral condition on the normal Laplacian of a submanifold, and proposes it as the right curvature equation for canonical foliations in codimension at least two. Its main theorem states that any smooth metric on $\\mathbb{R}^k \\times \\mathbb{S}^{n-k}$ sufficiently close to the product metric $g_0$ admits a unique smooth foliation by embedded $(n-k)$-spheres, each with QPMC and each close to some slice $\\{z\\} \\times \\mathbb{S}^{n-k}$. From this local result the paper builds a canonical normal form for bubblesheets, the high-curvature regions resembling $\\mathbb{R}^k \\times \\mathbb{S}^{n-k}$ that arise near singularities of mean curvature flow and Ricci flow: such a region is covered by a unique foliation of nearly vertical QPMC spheres. The significance is that these regions, previously described only up to approximate models, now carry distinguished geometrically defined leaves, analogous to the CMC neck foliations used in geometric flow surgery.","feed_headline":"Near-product metrics get a unique foliation by almost-round spheres","feed_subtitle":"A new curvature condition, QPMC, gives bubblesheet singularities a canonical normal form in geometric flows.","key_machinery":"The central object is the operator $Q$, the $L^2$-orthogonal projection of the normal bundle onto the sum of the eigenspaces of the normal Laplacian $-\\Delta^\\perp$ with eigenvalues below $\\lambda_{k+1}$, where $k$ is the codimension. A submanifold has quasi-parallel mean curvature when $(1-Q)(H)=0$, meaning $H$ lies in this lowest $k$-dimensional spectral subspace. Near a totally geodesic sphere in $\\mathbb{R}^k \\times \\mathbb{S}^{n-k}$, $Q$ has constant rank $k$ and varies smoothly with the metric and the submanifold, so the QPMC equation becomes a weakly elliptic quasilinear system solvable by the implicit function theorem. This spectral projection is what makes the leaves canonical: it selects the section of the normal bundle that is as parallel as possible when no parallel sections exist.","core_discovery":"On the product $\\mathbb{R}^k \\times \\mathbb{S}^{n-k}$, the slices $\\{z\\} \\times \\mathbb{S}^{n-k}$ are totally geodesic. Deforming the metric slightly destroys them, but the paper proves they can be replaced by a unique stack of nearby spheres whose mean curvature vector lies in the lowest $k$ eigenspaces of the normal Laplacian, the QPMC condition. For codimension one this reduces to constant mean curvature; for parallel-mean-curvature submanifolds it is automatic, but it is strictly weaker, as the paper shows by metrics on $\\mathbb{R}^2 \\times \\mathbb{S}^1$ admitting QPMC foliations with no parallel-mean-curvature foliation. The proof runs through an implicit function theorem argument: the linearization of the QPMC equation at the product metric is the sphere Laplacian, invertible on mean-zero normal graphs, and a bootstrap using Schauder estimates upgrades solutions to smoothness. The result extends from the exact product to any $(\\varepsilon, L, n-k)$-cylindrical region, giving a global foliation by $\\delta$-vertical QPMC spheres whose leaves are unique among all such spheres meeting the region.","pith_inferences":["A natural extension, not pursued in the paper, is to replace the round sphere $\\mathbb{S}^{n-k}$ by any compact Einstein manifold with a spectral gap above its $k$-th eigenvalue; the same projection argument should produce canonical foliations of the corresponding model space.","If the center-of-mass map from the paper is applied to an isolated high-codimension system, the core $\\Gamma_t$ could play the role that CMC spheres play in defining center of mass, giving a canonical reference submanifold for such systems.","A concrete check of the QPMC mechanism: in the Berger-sphere example, the non-parallel leaves should have mean curvature supported only in the lowest $k$ spectral modes, so the failure of parallelism is invisible to the projection; computing these modes directly would test the construction.","For mean curvature flow, the foliation suggests a surgery rule for bubblesheet singularities in codimension $k \\geq 2$, extending the neck-detection and surgery pipeline; whether the leaves remain controlled under the flow is an open question the paper does not address."],"forward_implications":["Any metric $C^{\\ell-1,\\gamma}$-close to $g_0$ on $\\mathbb{R}^k \\times \\mathbb{S}^{n-k}$ is foliated by unique QPMC spheres, so the round slices deform canonically rather than arbitrarily.","In a bubblesheet region of a mean-curvature-flow solution, the paper produces a distinguished foliation by $\\delta$-vertical QPMC spheres, together with a center-of-mass map to an $m$-dimensional core whose curvature is much smaller than the local curvature scale.","If the whole hypersurface becomes a bubblesheet at the singular time, the enclosed region is isotopic to a tubular neighborhood of a closed $m$-dimensional submanifold, constraining the topology of the original hypersurface.","The examples on $\\mathbb{R}^2 \\times \\mathbb{S}^1$ show a QPMC foliation exists even when no foliation by parallel-mean-curvature spheres exists, so QPMC is the right level of generality for codimension-$k$ foliations.","The uniqueness statement makes the foliation canonical: any embedded $\\delta$-vertical QPMC sphere that meets the cylindrical region must be one of its leaves, so the construction does not depend on choices."],"supporting_citations":[{"why":"It establishes the CMC-foliation definition of center of mass that motivates canonical foliations and gives the codimension-one model.","marker":"[HY96]"},{"why":"It provides the precedent of a canonical maximal-hypersurface foliation used in a global stability proof.","marker":"[CK93]"},{"why":"It constructs the CMC neck foliation that the paper's bubblesheet normal form generalizes to higher codimension.","marker":"[Ham97]"},{"why":"It uses neck foliations to perform mean-curvature-flow surgery, the flow setting where bubblesheet regions form.","marker":"[HS09]"},{"why":"It supplies the Schauder estimates used in the regularity and implicit-function-theorem arguments.","marker":"[GM12]"},{"why":"It provides the Neck Detection Lemma converting pointwise curvature conditions into large epsilon-close cylindrical regions.","marker":"[HK17]"},{"why":"It gives stability of closed geodesics in Berger spheres, used to show QPMC leaves need not have parallel mean curvature.","marker":"[TU22]"},{"why":"It supplies generic bumpiness of metrics, used to produce metrics with QPMC but no parallel-mean-curvature foliation.","marker":"[Whi91]"}],"fun_headline_variants":["QPMC condition gives bubblesheets a canonical sphere foliation","New curvature condition QPMC foliates near-product metrics uniquely","Canonical QPMC foliation for high-curvature bubble regions","Bubblesheets admit unique foliations by almost-round QPMC spheres","QPMC yields normal form for bubble singularities in flows"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The proof that local foliations agree on overlaps assumes that a nearly vertical, quasi-parallel sphere passing through an overlap point lies completely inside the first local chart, so that the local uniqueness theorem can identify the two foliations; this containment is not proved from the stated estimates.","fun_headline_variants_meta":{"raw":{"variants":["QPMC condition gives bubblesheets a canonical sphere foliation","New curvature condition QPMC foliates near-product metrics uniquely","Canonical QPMC foliation for high-curvature bubble regions","Bubblesheets admit unique foliations by almost-round QPMC spheres","QPMC yields normal form for bubble singularities in flows"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000197,"raw_usage":{"total_tokens":1349,"prompt_tokens":911,"completion_tokens":438,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":527,"completion_tokens_details":{"reasoning_tokens":351}},"tokens_in":527,"tokens_out":438,"duration_ms":4680,"temperature":1.0,"reasoning_tokens":351,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T15:17:38.292480+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"In the $\\mathbb{R}^2 \\times \\mathbb{S}^1$ example of the paper, compute the eigenvalues of the normal Laplacian on the QPMC leaves as the metric approaches the product metric; if $\\lambda_{k+1}-\\lambda_k$ approaches zero on a sequence of leaves, the projection $Q$ stops having constant rank $k$ and the frame construction behind the implicit-function-theorem proof breaks down, contradicting the claimed uniform foliation near that limit.","supporting_citations":[],"review_version":1}