{"id":"99faf04c-60cd-4675-bf0a-c1ce1b9cc881","arxiv_id":"2607.27711","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":8.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Inverse mean curvature flow produces 1-Lipschitz measure-preserving maps from the round sphere onto every smooth two-sphere of Gaussian curvature at least one, resolving E. Milman's contraction conjecture in dimension two.","lead":"This paper uses inverse mean curvature flow to build area-preserving contraction maps from the round sphere or flat disk onto convex hypersurfaces, proving E. Milman's contraction conjecture for Riemannian two-spheres with curvature at least one. The maps also yield two-sided spectral inequalities and a new map-level rigidity result, with an analogous disk theorem for nonnegatively curved free-boundary surfaces.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The endpoint construction depends on [25, Thm 1.4] and [19, Thm 1.1] providing C^{1,β} radial-graph convergence of IMCF for every strictly convex initial hypersurface; if those theorems have hidden hypotheses (e.g., star-shapedness), Theorem 1.1 is not established.","rationale":"The paper's novel contribution is to reduce E. Milman's conjecture to the construction of contraction maps via IMCF. The reduction itself is internally coherent: Proposition 2.4 correctly turns metric/volume monotonicity into a contraction, and the rigidity proof in Theorem 4.5 is a substantial and apparently correct argument. However, the entire extrinsic construction rests on the assertion that IMCF converges to a totally geodesic equator (or flat disk) in the C^{1,β} graph sense, with the specific radial-graph parametrization used in §3 and §5. This is cited to [25] and [19] but not restated. If the cited theorems are weaker or more conditional than claimed — for instance, if they only cover star-shaped initial data — then the maps T_t may not converge to a uniform bi-Lipschitz limit, and the central theorem would fail. This is an external-verification risk rather than an internal mathematical error, and it aligns with the reader's weakest_assumption. Hence I maintain the CONDITIONAL verdict.","tokens_in":17138,"tokens_out":26456,"duration_ms":232613,"concrete_test":"Verify the exact statements and hypotheses of Makowski–Scheuer [25, Theorem 1.4] and Lambert–Scheuer [19, Theorem 1.1 and Remark 7.4]. In particular, confirm that for every smooth closed strictly convex hypersurface in S^{n+1} (and every strictly convex free-boundary disk-type hypersurface in B^{n+1}), the flow exists up to the terminal time with |Σ_t|→|S^n| (resp. ω_n) and that the graph functions u_t converge to π/2 (resp. 1) in C^{1,β} for some β>0. If either theorem requires the initial hypersurface to be star-shaped w.r.t. a specified point, check whether that condition follows from the hypotheses of Theorem 1.2; if not, the proof of Proposition 2.4 has a gap. A direct analytical or numerical check on an off-center convex cap in S^3 would settle the applicability.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim (Theorem 1.1) is deduced from the extrinsic contraction theorem 1.2(1), whose proof hinges on the asserted terminal convergence of inverse mean curvature flow: for a closed strictly convex hypersurface Σ0 ⊂ S^{n+1}, [25, Thm 1.4] is invoked to guarantee that, after some t0, Σt is a radial graph r=u_t(z) over a fixed equator with u_t → π/2 in C^{1,β} (eq. 3.2). This is used in two essential places: (i) the radial-graph form of the metric q_t = du_t^2 + sin^2(u_t) g_can in (3.5), and (ii) the uniform convergence q_t → g_can that enables the Arzelà–Ascoli/endpoint argument in Prop. 2.4. If [25] only gives convergence in a weaker smoothness class, or only for a subclass of convex hypersurfaces that are already star-shaped w.r.t. the flow's limit equator, then the maps T_t may not converge uniformly to a bi-Lipschitz homeomorphism T, and the Lipschitz/measure-preservation conclusions of Theorem 1.2(1) — and hence of Theorem 1.1 — are not established. The same dependence appears in the free-boundary case through [19, Thm 1.1/Rem 7.4]. Because the paper does not restate the hypotheses of these cited theorems, this is the least secure link in the proof. The internal flow/transport arguments (Sections 2 and Props. 2.1–2.4) and the rigidity proof (Thm 4.5) appear mathematically sound; the uncertainty is at the citation–hypothesis boundary.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper introduces a flow-based construction of normalized-area-preserving Lipschitz contractions. The central mechanism is that for a normal flow of convex hypersurfaces with inverse-mean-curvature speed, the backward flow maps are 1-Lipschitz and preserve normalized area (Props. 2.1--2.2). A compactness endpoint argument (Prop. 2.4) upgrades these finite-time backward maps to a bi-Lipschitz area-preserving contraction from the terminal model hypersurface to the initial one. This is applied, first, to inverse mean curvature flow of closed strictly convex hypersurfaces in S^{n+1} and, second, to free-boundary convex hypersurfaces in B^{n+1}, yielding Theorem 1.2. In dimension two, the closed case is combined with Lu's spherical Weyl embedding to prove Theorem 1.1, E. Milman's contraction conjecture for smooth Riemannian two-spheres with K_g>=1; the free-boundary case is combined with Koerber's Weyl theorem to prove the analogous intrinsic disk theorem (Thm. 1.3). The paper also establishes map-level spectral rigidity (Thm. 4.5) and recovers the Lin--Wang--Xu spectral rigidity as a corollary.","tokens_in":17438,"tokens_out":16752,"duration_ms":166652,"significance":"If correct, Theorem 1.1 settles the last open nontrivial case of E. Milman's conjecture: the normalized round sphere dominates every smooth positively curved two-sphere in Gromov's Lipschitz order. The method is original and elegant: inverse mean curvature flow is singled out from first-variation formulas as the unique normal flow whose backward maps satisfy both contraction and normalized-area preservation. The proof is essentially parameter-free and the chain of reductions is transparent: flow convergence -> endpoint map -> Weyl embedding -> intrinsic theorem. The main external inputs are independent published convergence theorems of Makowski--Scheuer and Lambert--Scheuer and Weyl embedding theorems; the text verifies the hypotheses it states. The rigidity theorem is a strong addition, giving equality cases at the actual Lipschitz threshold.","major_comments":[],"minor_comments":[{"comment":"The endpoint construction depends on the C^{1,β} radial-graph convergence from [25, Thm. 1.4]. The text cites Assumption 1.3(i) of [25] and the convex-body result [6], but does not reproduce the precise statement or full hypotheses. Since this is the load-bearing external input, please add a short remark stating exactly which hypotheses of [25, Thm. 1.4] are satisfied and what the theorem guarantees, especially the existence of a fixed limiting equator over which the flow is a radial graph from some time onward.","section":"§3, Eq. (3.2)"},{"comment":"The same request applies to [19, Thm. 1.1 and Rem. 7.4]: state the hypotheses already verified (including the one-sided condition from (5.2)) and the precise convergence statement used in (5.6). Also, the notation Q_t(x)=f(x,u_t(x)) is introduced without defining f; please either define it or give the precise reference to the Möbius graph parametrization of [19, §5], including the property f(x,1)=x.","section":"§5, around Eq. (5.3)"},{"comment":"The passage from the approximating maps T_ε to the limit T is summarized as 'the same argument as in the proof of Proposition 2.4'. Since the lower bound in (4.5) is what upgrades the uniform limit to a bi-Lipschitz homeomorphism, it would help to spell out the injectivity and surjectivity argument in one or two sentences, rather than relying on an analogy.","section":"§4, Proof of Theorem 1.1"},{"comment":"The citation [3, Cor. 2.14] for Rademacher's theorem on Alexandrov spaces may not be the standard source; the metric-measure calculus of [18] seems more directly applicable. Please check and correct the reference.","section":"§2, Lemma 2.3"},{"comment":"Minor typographical issues: 'Gronwall' should be 'Grönwall', 'F unding' in the acknowledgments section should be 'Funding', and the spelling of 'Möbius'/'Mobius' should be made consistent.","section":"Miscellaneous"}],"recommendation":"minor_revision","confidential_remarks":"The manuscript's central argument is sound to this referee. The main risk is the dependence on the terminal C^{1,β} convergence theorems [25] and [19]; I do not find a concrete error, but the authors should be asked to make the exact hypotheses and conclusions of these theorems explicit, since the endpoint construction in Prop. 2.4 carries the whole proof. The manuscript also relies on several 2026 arXiv preprints ([2], [7], [22]); the editor may wish to ensure these are in final or at least stable form."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear colleague,\n\nThe short version: this paper proves E. Milman's contraction conjecture for smooth Riemannian two-spheres with Gaussian curvature at least one, and it does it with an original, clean idea — inverse mean curvature flow's backward maps are, by first variation, the unique normal-flow maps that both contract distance and preserve normalized area. The endpoint compactness argument (Prop 2.4) turns the flow's terminal convergence into a bi-Lipschitz contraction onto the initial hypersurface. That is genuinely new, and it closes the dimension where the conjecture was still open.\n\nWhat the paper does well: the first-variation derivation in §2 is crisp; Prop 2.2 makes clear why IMCF is the right flow, and the co-Lipschitz estimate Lemma 2.3 is useful beyond this application. The rigidity result, Theorem 4.5, is substantive — it applies to any area-preserving Lipschitz map from the round sphere to a smooth two-sphere, shows equality in the spectral lower bound forces a=L^2 and isometry, and recovers Lin–Wang–Xu's spectral rigidity. That part is self-contained and convincing. The Alexandrov and free-boundary extensions follow the same template, and the conformal-approximation arguments look fine.\n\nThe soft spot, as flagged in the stress-test note, is the endpoint construction's dependence on external convergence theorems: [25, Thm 1.4] for closed IMCF in S^{n+1} and [19, Thm 1.1/Rem 7.4] for free-boundary IMCF. The paper states the resulting C^{1,β} radial-graph convergence to the equator/flat disk but does not restate the hypotheses of those theorems. If those theorems carry extra assumptions (e.g. the initial hypersurface is already star-shaped, or convergence is in a weaker norm), the uniform metric convergence q_t→g_can and hence the Arzelà–Ascoli step in Prop 2.4 would need adjustment. This is not a flaw I can confirm from the text — the citations look appropriate, and the flow literature does have such results for convex hypersurfaces — but it is the load-bearing link that a referee must check. The same applies to Lu's spherical Weyl embedding and Koerber's free-boundary Weyl theorem, on which the intrinsic theorems depend.\n\nOverall: the internal machinery is sound, the result is important, and the citation-boundary uncertainty is exactly what peer review is for. I'd send it out. The referee should be asked to confirm the stated convergence theorems cover every smooth strictly convex initial hypersurface, and whether the regularity upgrade (Lemma 4.1 via Pogorelov) is needed at the stated level.\n\nWho it's for: anyone working on optimal transport, metric geometry, spectral comparison, or curvature flows. I'd cite it if the endpoint verification pans out.\n\nRecommendation: accept for peer review.","headline":"Solves the last open dimension of Milman's contraction conjecture via an original IMCF construction; the internal argument is clean and the main caveat is a stack of external convergence/embedding theorems that referees should verify.","tokens_in":18051,"tokens_out":2526,"would_cite":true,"duration_ms":24319,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["53C21","53E10","49Q22","53C45","58J50"],"pacs":[],"model":"deepseek-v4-flash","headline":"Inverse mean curvature flow builds normalized-area-preserving Lipschitz contractions that carry the round sphere onto every smooth positively curved two-sphere, proving the contraction conjecture in the last nontrivial dimension.","keywords":["inverse mean curvature flow","contraction maps","Riemannian two-spheres","Lipschitz order","free-boundary hypersurfaces","spectral comparison","Alexandrov surfaces","optimal transport"],"falsifier":"Measure the infimum Lipschitz constant among area-preserving maps from the round sphere to a positively curved two-sphere; any metric for which this infimum is greater than one would refute the theorem. A useful test is a sequence of smooth metrics approaching a diameter-π football-shaped limit: if the limiting infimum exceeds 1, the main claim fails.","tokens_in":16889,"feed_emoji":"🔵","tokens_out":11244,"duration_ms":109892,"temperature":0.7,"pith_summary":"Working with the inverse mean curvature flow of a strictly convex hypersurface, this paper proves that the backward flow maps are automatically 1-Lipschitz and preserve normalized area, because the metric grows while the area form scales uniformly. Sending the flow to its terminal model (the equator in the sphere, the flat disk in the ball) yields a concrete bi-Lipschitz contraction from the model onto the initial surface. In dimension two this establishes the contraction conjecture: every smooth Riemannian two-sphere with Gaussian curvature at least one is the image of the round sphere under such an area-preserving map with Lipschitz constant at most one, and strictly less than one when the curvature is strictly positive. The same construction gives contractions from the Euclidean ball to strictly convex free-boundary hypersurfaces and to intrinsically nonnegatively curved disks with boundary geodesic curvature one. Consequences include two-sided spectral comparison with rigidity: any equality in the lower spectral bound forces the target metric to be round.","feed_headline":"Round sphere contracts onto every positively curved 2-sphere","feed_subtitle":"Inverse mean curvature flow yields 1-Lipschitz area-preserving maps, unlocking spectral rigidity on 2-spheres.","key_machinery":"Inverse mean curvature flow (IMCF): the evolution of a hypersurface with normal speed equal to the reciprocal of its mean curvature. Its load-bearing identity is the pair of first-variation formulas ∂_t g_t = 2H^{-1} h_t and ∂_t dA_t = dA_t, which translate into metric growth and uniform area growth. The paper's endpoint proposition then passes from finite-time flow maps to the terminal limit, using the terminal convergence of IMCF to the equator (closed case) or the flat disk (free-boundary case) to produce the final contraction.","core_discovery":"The paper's central discovery is that inverse mean curvature flow is the unique positive normal flow whose backward Lagrangian maps are both distance-contracting and normalized-area-preserving. Under the speed f=H^{-1}, the induced metric satisfies ∂_t g_t = 2H^{-1} h_t ≥ 0 while the area form satisfies ∂_t dA_t = dA_t; the first inequality means backward maps shrink distances, and the spatially constant area growth means they preserve normalized area. By proving a compactness result that sends these finite-time maps to the flow's terminal limit, the authors obtain a bi-Lipschitz homeomorphism from the terminal model (round equator or flat disk) onto any smooth strictly convex initial hypers","pith_inferences":["If the main construction is robust, the same endpoint-contraction principle should work for other curvature flows with speeds satisfying fH=1 after reparametrization; this suggests a general recipe for flow-generated transport maps with prescribed Lipschitz constants.","For rotationally symmetric targets the IMCF evolution is explicit, so the Lipschitz constant of the constructed map should be computable; comparing it with diameter or width could yield a quantitative 'roundness modulus' and sharpen the rigidity statement before spectral equality.","The appearance of equality only in the round/large-sphere limit and in singular Alexandrov footballs suggests the boundary of the contraction theorem is marked by metric degenerations; testing a sequence of smooth metrics converging to a football would clarify whether the Lipschitz constant tracks the metric distance to the round sphere.","In the free-boundary setting the extrinsic theorem already holds in all dimensions, so a higher-dimensional intrinsic disk theorem would follow if a free-boundary embedding theorem analogous to the two-dimensional one is available; this is a concrete open direction."],"forward_implications":["Every smooth two-sphere (M,g) with Gaussian curvature at least 1 admits a normalized-area-preserving bi-Lipschitz contraction T:S²→M with Lip(T)≤1; consequently the normalized round sphere dominates (M,g) in the metric–measure Lipschitz order, transferring isoperimetric, concentration, and spectral inequalities to M.","Two-sided spectral comparison holds with the actual Lipschitz constant: L^{-2}λ_k(S²,gcan) ≤ λ_k(M,g) ≤ (L/a)²λ_k(S²,gcan), and equality for any k≥1 forces T to be an isometry after scaling, so the only equality metric is round.","For closed strictly convex hypersurfaces in the sphere, the contraction has Lipschitz constant strictly below one; for geodesic spheres the bound is sharp, with equality in the two-sided estimate, showing the strictness cannot be uniform.","For strictly convex free-boundary hypersurfaces in the unit ball, the flat disk contracts onto Σ with L<1, giving |Σ|<ω_n and λ^{D/N}_k(Σ) ≥ L^{-2}λ^{D/N}_k(B^n).","The intrinsic disk theorem extends the same conclusions to smooth metrics on the disk with Kg≥0 and boundary geodesic curvature 1, yielding spectral comparison for Dirichlet and Neumann Laplacians."],"fun_headline_variants":["IMCF builds contractions onto every convex shape","Milman's contraction conjecture proven by IMCF","1-Lipschitz area-preserving maps via IMCF","Inverse mean curvature flow: shrink distance, keep area"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The whole argument depends on the flow's terminal behavior: inverse mean curvature flow must settle smoothly onto the equator (or flat disk), and every intrinsic metric satisfying the curvature bounds must be realizable as a strictly convex surface; if either fails, the claimed contraction maps are not guaranteed.","fun_headline_variants_meta":{"raw":{"variants":["IMCF builds contractions onto every convex shape","Milman's contraction conjecture proven by IMCF","1-Lipschitz area-preserving maps via IMCF","Inverse mean curvature flow: shrink distance, keep area"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001165,"raw_usage":{"total_tokens":4609,"prompt_tokens":649,"completion_tokens":3960,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":393,"completion_tokens_details":{"reasoning_tokens":3897}},"tokens_in":393,"tokens_out":3960,"duration_ms":31714,"temperature":1.0,"reasoning_tokens":3897,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-01T02:50:01.103360+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the infimum Lipschitz constant among area-preserving maps from the round sphere to a positively curved two-sphere; any metric for which this infimum is greater than one would refute the theorem. A useful test is a sequence of smooth metrics approaching a diameter-π football-shaped limit: if the limiting infimum exceeds 1, the main claim fails.","supporting_citations":[],"review_version":1}