{"id":"ee5855e9-2e55-4e19-89e2-e6d3a041de12","arxiv_id":"2504.15109","paper_version":1,"verdict":"ACCEPT","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"low","formal_verification":"none","parameter_count":0,"one_line_summary":"A Heintze-Karcher inequality is proved for hyperbolic domains with mean curvature greater than -1, plus shifted-curvature versions in sub-static warped products.","lead":"This paper proves new Heintze-Karcher type inequalities for hypersurfaces with shifted mean curvature in hyperbolic space and in sub-static warped product manifolds. The main novelty is a bound that works for non-mean-convex domains in hyperbolic space, with applications that force solutions of curvature equations to be spheres.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Equality case of Theorem 1.1 is underproved: Q(0)=0 with Q(t)≤0 and liminf Q(t)≥0 does not force Q(t)=0 without showing the auxiliary domains {w>t} satisfy the same inequality.","rationale":"The reader's verdict ACCEPT is reasonable for the inequality part: Lemma 3.1's differential estimate and the algebraic derivation of Theorem 1.4 check out, and Lemma 3.4 is not a genuine weak spot—it follows from the divergence theorem applied to V=sinh r ∂_r on {w>t}, since the outward normal is -∇w and the cut locus has measure zero. The real soft spot is the equality case of Theorem 1.1. The proof shows Q(0)≥Q(t) and liminf Q(t)≥0; equality in the theorem gives only Q(0)=0, hence Q(t)≤0 and liminf Q(t)≥0, which does not imply Q(t)=0 unless one independently knows Q(t)≥0. The single-sentence appeal to Lemma 3.1 omits the necessary argument that each auxiliary domain {w>t} satisfies the Heintze-Karcher inequality via smooth approximation, which would force Q(t)=0 and propagate the umbilic condition. This gap is load-bearing because Theorem 1.2's uniqueness conclusion relies on the equality case. It is likely fixable, so CONDITIONAL rather than REJECT is appropriate.","tokens_in":15525,"tokens_out":33696,"duration_ms":287301,"concrete_test":"Complete the equality-case proof by showing Q(t)≥0 for all t∈[0,T): for each t, approximate {w>t} by smooth domains Ω_{t,j} whose boundaries lie in a small tubular neighborhood of Σ_t^*∪C, apply the already-proven inequality (1.4) to each Ω_{t,j}, and pass to the limit j→∞. If Q(t)≥0, then equality in (1.4) gives Q(t)=0 for all t, and Lemma 3.1 implies each Σ_t^* is umbilic, hence Σ is umbilic. If this approximation argument fails (e.g., the inequality is not continuous under the chosen approximation), the equality statement of Theorem 1.1 is unsupported.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The main inequality in Theorem 1.1 is established via a monotonicity argument: Q(0)≥Q(t) for all t∈[0,T) and liminf_{t→T} Q(t)≥0. If equality holds in (1.4), then Q(0)=0, so one gets Q(t)≤0 for all t and liminf Q(t)≥0. These conditions do not force Q(t)=0: Q(t) could be negative on [0,T) and approach 0 as t→T. The final sentence 'If equality holds in (1.4), then Lemma 3.1 implies that Σ is umbilic' therefore skips a step. To use Lemma 3.1's equality condition, one must know that equality propagates along the flow, i.e., that Q(t)=0 for almost every t. This requires first showing that each approximate domain {w>t} satisfies the same Heintze-Karcher inequality, so Q(t)≥0, which entails a smooth approximation of the non-smooth boundary {w>t}∪C (the cut locus has measure zero but is not smooth). The paper does not supply this argument. Because the equality case is used in the proof of Theorem 1.2 via the chain of inequalities in Section 5, this is a genuine gap, although likely repairable by a standard approximation argument.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proves new Heintze-Karcher type inequalities involving the shifted mean curvature. Theorem 1.1 states that for a bounded domain Ω with smooth boundary Σ in hyperbolic space H^{n+1}, if the mean curvature p_1(κ) satisfies p_1(κ)>-1, then ∫_Σ (λ'+u)/(p_1(κ)+1) dμ ≥ (n+1)∫_Ω λ' dv, with equality iff Σ is umbilic and hence a geodesic sphere. Theorem 1.4 extends this to sub-static warped product manifolds under a static-convexity condition and a sign condition for a general shift ε, yielding inequalities (1.12) and (1.13). As applications, the authors prove uniqueness theorems for hypersurfaces satisfying curvature equations (Theorems 1.2 and 1.6) and an Alexandrov-type corollary. The proofs combine the inward unit normal flow, a monotonicity argument, Minkowski formulas, and a generalized Reilly formula from [17].","tokens_in":15772,"tokens_out":23738,"duration_ms":193615,"significance":"If correct, Theorem 1.1 is a meaningful advance: it extends the Heintze-Karcher inequality to domains whose boundary is not assumed mean-convex, which is new in this setting. Theorem 1.4 provides a unified framework that recovers known results for shifted factors ε=0,±1 and extends them to arbitrary ε under convexity hypotheses. The applications to shifted curvature equations are natural and yield clean uniqueness statements. The flow computation in Lemma 3.1 is carefully executed, and the derivation of Theorem 1.4 from Theorem B is algebraic and transparent. The main shortcomings are the underproved equality case of Theorem 1.1 and a technical gap in the monotonicity argument; these are repairable but affect the stated applications for k=1.","major_comments":[{"comment":"The equality case of Theorem 1.1 is not proven. From equality in (1.4) one obtains Q(0)=0. The monotonicity established in Lemma 3.5 gives Q(0)≥Q(t) for all t, and the lower bound (3.7) gives liminf_{t→T} Q(t)≥0. These two facts together do not imply Q(t)=0 for any t: Q(t) could be strictly negative on [0,T) and approach 0 as t→T. Therefore the sentence 'If equality holds in (1.4), then Lemma 3.1 implies that Σ is umbilic' skips a necessary step: to apply Lemma 3.1's equality condition one must know that equality propagates along the flow, e.g., that Q(t)=0 for almost every t. This would follow if each approximate domain {w>t} satisfied the same Heintze-Karcher inequality (giving Q(t)≥0), but the boundary {w>t}∪C is not smooth and no approximation argument is supplied. Since this equality case is used in the proof of Theorem 1.2 for k=1 (via the chain in Section 5), the gap is load-bearing. The authors should either prove Q(t)≥0 for a.e. t by a smoothing argument or supply an alternative proof of the equality statement.","section":"Section 3, proof of Theorem 1.1 (equality case)"},{"comment":"The passage from the pointwise limsup inequality limsup_{h↘0} h^{-1}(Q(t)-Q(t-h)) ≤ (n+1)e^{-(n+1)t}((n+1)∫_{w>t}λ'dv - ∫_{Σ*_t}u dμ_t) to the integrated bound Q(t)-Q(0) ≤ ∫_0^t (n+1)e^{-(n+1)τ}((n+1)∫_{w>τ}λ'dv - ∫_{Σ*_τ}u dμ_τ)dτ requires a justification that Q is absolutely continuous, or at least a comparison lemma for Dini derivatives valid under the one-sided, almost everywhere condition at hand. The function Q involves integrals over the level sets Σ*_t, which are smooth only away from the cut locus; the paper does not explain why differentiation under the integral is legitimate in the presence of the cut locus. This is a technical gap in the proof of the main inequality itself, although it appears repairable by a standard coarea/approximation argument.","section":"Section 3, Lemma 3.5"}],"minor_comments":[{"comment":"In the definition of the set A, the expression 'A={(x,t)∈Σ∈[0,∞): w(X(x,t))=t}' should read 'A={(x,t)∈Σ×[0,∞): w(X(x,t))=t}'.","section":"Section 3"},{"comment":"The displayed ODE 'd/dt log ξ(t) = ε/((1-ε^2)λ'-ε t)' appears to contain a typo in the denominator; as written it mixes a function λ'(r) with the flow parameter t in a dimensionally inconsistent way.","section":"Remark 3.2"},{"comment":"The statement 'equality holds in (4.7) if and only if p_1(κ) is constant on Σ' is slightly imprecise: equality in the Cauchy-Schwarz step also requires that λ'-εu does not change sign on Σ. Equality in (1.12) forces this sign condition, but the proof does not mention this.","section":"Section 4, equation (4.7)"},{"comment":"The paper cites Lemma 3.4 from [15] without reproducing its proof; since this lemma is crucial for the monotonicity of Q, a brief indication of its origin and why the cut locus does not affect it would improve readability.","section":"Throughout"}],"recommendation":"major_revision","confidential_remarks":"This is a solid paper with a likely correct main inequality. The equality-case gap and the technical issue in the monotonicity step are repairable and do not appear to undermine the central inequality, but they do affect the stated applications for k=1. I would support publication after a major revision addressing these points."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper’s main contribution is Theorem 1.1: a Heintze–Karcher inequality for domains in hyperbolic space whose mean curvature is only bounded below by −1, not by 0 or 1. That genuinely removes a long-standing convexity restriction, and the flow computation in Lemma 3.1 closes because ε=−1, F=−1 makes the extra term T vanish. This is a real piece of new geometry, not a repackaging. Theorem 1.4 is more modest: it is a clean algebraic derivation of a shifted Heintze–Karcher inequality from Li–Xia’s Minkowski inequality under a static-convexity assumption. That is still useful because it unifies the known ε=0, ±1 cases and states the hypothesis explicitly, but it is not a new flow argument.\n\nThe soft spot is the equality case of Theorem 1.1. The proof of the inequality itself is fine: Q(0) ≥ Q(t) and liminf Q(t) ≥ 0 give the desired bound. But if equality holds, the argument only yields Q(0)=0 and hence Q(t) ≤ 0; it does not force Q(t)=0 along the flow. To invoke the equality condition in Lemma 3.1, one would need to know that each approximate domain {w>t} satisfies the same Heintze–Karcher inequality, so that Q(t) ≥ 0, and that requires a smooth approximation argument for the non-smooth level sets. The paper skips this in one sentence. The gap is real and it matters, because Theorem 1.2 uses the equality case of Theorem 1.1 to conclude that solutions are geodesic spheres. The gap is probably repairable by a standard approximation argument, but it is not present in the manuscript. Also, Lemma 3.4 is cited rather than proved; that is acceptable if the cited statement is correct, but it does make Theorem 1.1 conditional on a cut-locus identity that the reader cannot check without going to the source.\n\nThe algebraic derivation of Theorem 1.4 is sound, and the paper is honest about the static-convexity limitation in Problem 1.1. The applications are meaningful and the references look appropriate; self-citations here are used as context, not as circular inputs.\n\nMy recommendation: send this to a serious referee. The main inequality is new and the flow argument is largely correct. The referee should be asked to scrutinize the equality case and to require the missing approximation step before the paper is accepted.","headline":"The new non-mean-convex Heintze–Karcher inequality in hyperbolic space is real and the flow computation is sound, but the equality case of Theorem 1.1 is underproved and the applications inherit that gap.","tokens_in":16354,"tokens_out":2600,"would_cite":true,"duration_ms":24460,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["53C42","53C24","53C21"],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper proves Heintze-Karcher type inequalities for shifted mean curvature in sub-static warped product manifolds, including a new hyperbolic-space bound for domains whose boundary mean curvature is merely greater than -1, with…","keywords":["Heintze-Karcher inequality","shifted principal curvatures","sub-static warped product","hyperbolic space","unit normal flow","geodesic sphere rigidity","curvature equations","weighted integral inequality"],"falsifier":"Numerically evaluate $\\mathcal{I}=\\int_\\Sigma \\frac{\\lambda'+u}{p_1(\\kappa)+1}\\,d\\mu-(n+1)\\int_\\Omega \\lambda'\\,dv$ for a smooth, non-umbilic surface of revolution in $\\mathbb{H}^3$ whose mean curvature is everywhere $>-1$ but that has a negative principal curvature somewhere; if any such surface gives $\\mathcal{I}<0$, Theorem 1.1 is false, whereas the theorem predicts $\\mathcal{I}\\ge0$ with equality only for geodesic spheres.","tokens_in":15285,"feed_emoji":"📐","tokens_out":11441,"duration_ms":89884,"temperature":0.7,"pith_summary":"The paper establishes new Heintze-Karcher type integral inequalities in which the boundary integrand uses the shifted mean curvature $p_1(\\kappa)-\\varepsilon$ rather than the usual mean curvature. Its main new result is a hyperbolic-space inequality that does not require the boundary to be mean-convex: if $\\Omega\\subset\\mathbb{H}^{n+1}$ has smooth boundary $\\Sigma$ with $p_1(\\kappa)>-1$, then $\\int_\\Sigma \\frac{\\lambda'+u}{p_1(\\kappa)+1}\\,d\\mu \\ge (n+1)\\int_\\Omega \\lambda'\\,dv$, with equality only for umbilic boundaries, i.e. geodesic spheres. For general sub-static warped product manifolds, the paper proves an analogous inequality for every shift $\\varepsilon\\in\\mathbb{R}$ under a static-convexity hypothesis and a positivity condition, and uses these estimates to show that certain curvature equations force hypersurfaces to be geodesic spheres. This matters because shifted mean curvature is the natural curvature in hyperbolic geometry, and the new inequalities extend a tool that is widely used to prove rigidity and isoperimetric results.","feed_headline":"Hyperbolic volume bound holds when mean curvature exceeds -1","feed_subtitle":"Weighted boundary integral dominates weighted volume; equality only on geodesic spheres.","key_machinery":"The central objects are the shifted principal curvatures $\\tilde\\kappa_i=\\kappa_i-\\varepsilon$ and the shifted mean curvature $p_1(\\kappa)-\\varepsilon$. The carrying identity is the shifted integral formula $\\int_\\Sigma(\\lambda'-\\varepsilon u)\\,p_{m-1}(\\kappa-\\varepsilon)\\,d\\mu=\\int_\\Sigma u\\,p_m(\\kappa-\\varepsilon)\\,d\\mu$ (Lemma 2.5), derived from the Hessian of $\\Phi(r)$ and the Codazzi property. For Theorem 1.1, the mechanism is the inward unit normal flow $\\partial_t X=-\\nu$, the evolution inequality of Lemma 3.1, and the auxiliary functional $Q(t)=e^{-(n+1)t}\\left(\\int_{\\Sigma_t^*}\\frac{\\lambda'+u}{p_1(\\kappa)+1}\\,d\\mu-(n+1)\\int_{\\{w>t\\}}\\lambda'\\,dv\\right)$, whose monotonicity follows from a quoted almost-everywhere identity for the support function and yields the global bound by taking $t\\to T$. For Theorem 1.4, the mechanism is an algebraic rearrangement of a weighted integral inequality into a Cauchy-Schwarz comparison, reducing the shifted Heintze-Karcher inequality to the static-convex case.","core_discovery":"The central claim is Theorem 1.1: in $\\mathbb{H}^{n+1}$, any bounded domain whose boundary satisfies $p_1(\\kappa)>-1$ obeys $\\int_\\Sigma \\frac{\\lambda'+u}{p_1(\\kappa)+1}\\,d\\mu \\ge (n+1)\\int_\\Omega \\lambda'\\,dv$, where $\\lambda'=\\cosh r$ and $u=\\langle \\sinh r\\,\\partial_r,\\nu\\rangle$ is the support function of $\\Sigma$; equality holds precisely when $\\Sigma$ is a geodesic sphere. The threshold $-1$ is what makes the statement new, since it allows boundaries that are not mean-convex. For a general sub-static warped product $\\bar g=dr^2+\\lambda(r)^2g_N$ with potential $\\lambda'$, Theorem 1.4 states that if $\\Sigma$ is static-convex and $(\\lambda'-\\varepsilon u)(p_1(\\kappa)-\\varepsilon)>0$, then $\\int_\\Sigma \\frac{\\lambda'-\\varepsilon u}{p_1(\\kappa)-\\varepsilon}\\,d\\mu \\ge (n+1)\\int_\\Omega \\lambda'\\,dv$, with a boundary correction when the manifold has a horizon; the equality cases are those of the underlying weighted integral inequality. The paper treats the three cases $\\varepsilon=0$, $\\varepsilon=1$, and $\\varepsilon=-1$ as special instances of this general principle.","pith_inferences":["Not claimed in the paper: the static-convexity question raised in Problem 1.1 could be tested numerically with non-static-convex, shifted convex surfaces in $\\mathbb{H}^3$.","Not claimed in the paper: Theorem 1.4's derivation depends only on an abstract weighted integral inequality, so the shifted bound should transfer to any ambient manifold admitting such an inequality.","Not claimed in the paper: the $p_1(\\kappa)>-1$ threshold points toward rigidity theorems for hyperbolic hypersurfaces with mean curvature bounded below by any constant above $-1$."],"forward_implications":["Every bounded domain in $\\mathbb{H}^{n+1}$ with boundary mean curvature above $-1$ satisfies (1.4); this removes the usual mean-convexity requirement from the Heintze-Karcher bound.","Theorem 1.4 gives one inequality that specializes to the $\\varepsilon=0$, $\\varepsilon=1$, and $\\varepsilon=-1$ shifted Heintze-Karcher inequalities in the relevant ambient spaces.","Closed hypersurfaces in $\\mathbb{H}^{n+1}$ solving $p_k(\\kappa+1)=\\chi(\\lambda',-\\lambda'-u)$ with $\\partial_1\\chi\\le0$, $\\partial_2\\chi\\ge0$, and the appropriate convexity are geodesic spheres (Theorem 1.2).","A constant shifted $k$-th mean curvature $p_k(\\kappa+1)$ on a closed hypersurface in $\\mathbb{H}^{n+1}$ forces the hypersurface to be a geodesic sphere (Corollary 1.3).","The same rigidity mechanism works in space forms for $p_k(\\kappa-\\varepsilon)=\\chi(\\Phi,\\varepsilon\\Phi-u)$ under static-convexity (Theorem 1.6)."],"supporting_citations":[{"why":"Supplies the almost-everywhere identity for the support function used in Lemma 3.5, the step that converts the flow estimate into the global inequality.","marker":"[15]"},{"why":"Introduces the unit normal flow in warped products and proves the epsilon=0 Heintze-Karcher inequality that this paper generalizes.","marker":"[4]"},{"why":"Provides the weighted integral inequality (Theorem B) from which Theorem 1.4 is derived by algebraic rearrangement.","marker":"[17]"},{"why":"Establishes the original Heintze-Karcher volume comparison that motivates the shifted inequalities.","marker":"[12]"},{"why":"Gives the Euclidean Heintze-Karcher inequality used as the model for the proof strategy.","marker":"[24]"},{"why":"Proves the n=1 case of the shifted inequality and introduces the horospherical setting that the applications build on.","marker":"[16]"}],"fun_headline_variants":["New HK inequality allows non-convex hyperbolic domains","Mean curvature > -1 suffices for Heintze-Karcher in H^n","Heintze-Karcher extends to non mean-convex hyperbolic sets","Boundary curvature bound lowered to -1 for weighted volume","Weighted volume bound holds for hyperbolic domains with p1 > -1"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The proof of the main hyperbolic inequality relies on a quoted almost-everywhere identity that relates an integral of the support function over a level set to a weighted volume of the region beyond it; if that identity fails, the monotonicity argument for Q(t) collapses and Theorem 1.1 does not follow.","fun_headline_variants_meta":{"raw":{"variants":["New HK inequality allows non-convex hyperbolic domains","Mean curvature > -1 suffices for Heintze-Karcher in H^n","Heintze-Karcher extends to non mean-convex hyperbolic sets","Boundary curvature bound lowered to -1 for weighted volume","Weighted volume bound holds for hyperbolic domains with p1 > -1"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000611,"raw_usage":{"total_tokens":2812,"prompt_tokens":883,"completion_tokens":1929,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":499,"completion_tokens_details":{"reasoning_tokens":1838}},"tokens_in":499,"tokens_out":1929,"duration_ms":11403,"temperature":1.0,"reasoning_tokens":1838,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-16T11:33:33.337377+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Numerically evaluate $\\mathcal{I}=\\int_\\Sigma \\frac{\\lambda'+u}{p_1(\\kappa)+1}\\,d\\mu-(n+1)\\int_\\Omega \\lambda'\\,dv$ for a smooth, non-umbilic surface of revolution in $\\mathbb{H}^3$ whose mean curvature is everywhere $>-1$ but that has a negative principal curvature somewhere; if any such surface gives $\\mathcal{I}<0$, Theorem 1.1 is false, whereas the theorem predicts $\\mathcal{I}\\ge0$ with equality only for geodesic spheres.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the almost-everywhere identity for the support function used in Lemma 3.5, the step that converts the flow estimate into the global inequality."},{"cited_title":"Brendle,Constant mean curvature surfaces in warped product manifolds, Publications math´ ematiques de l’IH´ES117(2013), 247–269","cited_arxiv_id":null,"evidence_quote":"Introduces the unit normal flow in warped products and proves the epsilon=0 Heintze-Karcher inequality that this paper generalizes."},{"cited_title":"Li and C","cited_arxiv_id":null,"evidence_quote":"Provides the weighted integral inequality (Theorem B) from which Theorem 1.4 is derived by algebraic rearrangement."},{"cited_title":"Heintze and H","cited_arxiv_id":null,"evidence_quote":"Establishes the original Heintze-Karcher volume comparison that motivates the shifted inequalities."},{"cited_title":"Ros,Compact hypersurfaces with constant higher order mean curvatures, Rev","cited_arxiv_id":null,"evidence_quote":"Gives the Euclidean Heintze-Karcher inequality used as the model for the proof strategy."}],"review_version":1}