Some rigidity results related to the Obata type equation

Yihu Yang; Yiwei Liu

REVIEW 3 major objections 4 minor 20 references

Some rigidity results related to the Obata type equation

T0 review · 3 major / 4 minor · reviewed 2026-08-12 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read A manifold with Ricci curvature at least $-n$ and boundary mean curvature at least $c>1$ must be a hyperbolic geodesic ball if it admits a non-constant solution of the Obata-type Robin equation.

desk verdict Useful new rigidity results for the Obata equation with Robin boundary condition c>1; the main arguments are sound but one load-bearing step is an unstated external theorem. read the letter →

arxiv 2411.18508 v2 pith:JNHTUT6A submitted 2024-11-27 math.DG

Yiwei Liu , Yihu Yang This is my paper · ORCID

classification math.DG MSC 53C2453C21

keywords ObatatypeequationRobinboundaryconditionhyperbolicspacerigiditytheoremwarpedproductmeancurvaturestandardsphereeigenvalueinequality

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper studies the Obata-type equation $\nabla^2 f - f g = 0$ with Robin boundary condition $f_\nu = c f$, $c = \coth\theta > 1$, on a complete Riemannian manifold with compact boundary. Its central claim is that, under the curvature bounds $\mathrm{Ric} \ge -n$ and boundary mean curvature $H \ge c$, the existence of a non-constant solution forces the manifold to be isometric to a geodesic ball of radius $\tanh^{-1}(1/c)$ in hyperbolic space $\mathbb{H}^{n+1}$. A companion theorem classifies the boundary under a different curvature condition plus a diameter lower bound, giving warped products over round spheres, and a spherical analogue forces a geodesic ball of radius $\tan^{-1}(1/c)$ in $\mathbb{S}^{n+1}$. The paper thus adds a new rigidity case to the Obata-type classification for Robin boundary conditions with $c>1$, where the solution has no critical points.

What carries the argument

The engine is the warped-product splitting induced by the level sets of $f$. From $\nabla^2 f = f g$ and $f_\nu=c f$ one obtains the constant $|\nabla f|^2 - f^2 = A>0$; after scaling $A=1$, the function $f$ becomes $\sinh t$ along its normalized gradient flow, whose integral curves are geodesics. The zero set $\Omega_0$ is a totally geodesic hypersurface, and $\Omega$ is embedded in the warped product $\Omega_0\times(-\infty,\infty)_t$ with metric $dt^2+\cosh^2 t\, g|_{\Omega_0}$ and boundary given by graphs $\pm\varphi$. The graph function $\varphi$ is constrained by $\cosh\varphi/\sqrt{1+\cosh^{-2}\varphi|\nabla^{\Omega_0}\varphi|^2}=c\sinh\varphi$, a PDE whose comparison with distance functions in $\mathbb{H}^n$ is what turns the curvature assumptions into the conclusion that $\Omega_0$ is a hyperbolic ball.

What would settle it

Take the model case of a hyperbolic geodesic ball of radius $\tanh^{-1}(1/c)$ with $f=\sinh t$ and check whether the two geometric assertions hold by direct computation. More decisively, build a complete warped product $\Omega_0\times\mathbb{R}_t$ with metric $dt^2+\cosh^2 t\, g|_{\Omega_0}$ over a compact base, choose a boundary graph $\varphi$ solving the paper's first-order equation, and test whether the first-exit curve orthogonality can fail; a single such example satisfying $\mathrm{Ric}\ge -n$ and $H\ge c$ but not isometric to the hyperbolic ball would refute Theorem 1.1.

Watch

Extended reading notes

Core claim

On its own terms, the paper establishes a rigidity classification. Let $\Omega^{n+1}$ be smooth, complete, connected, with compact boundary $\Sigma$, satisfying $\mathrm{Ric}_{\Omega}\ge -n$ and mean curvature $H\ge c=\coth\theta>1$. If a non-constant smooth $f$ solves $\nabla^2 f - f g = 0$ in $\Omega$ and $f_\nu = c f$ on $\Sigma$, then $\Omega$ is isometric to the geodesic ball of radius $\tanh^{-1}(1/c)$ in $\mathbb{H}^{n+1}$. The proof first normalizes the conserved quantity $|\nabla f|^2 - f^2$ to be $1$, so $f$ has no critical points, and uses the fact that flow lines of $\nabla f/|\nabla f|$ are geodesics. The zero level set $\Omega_0=\{f=0\}$ then carries a warped-product description of $\Omega$ as a $\mathbb{Z}_2$-symmetric domain in $\Omega_0\times\mathbb{R}_t$ with metric $dt^2 + \cosh^2 t\, g|_{\Omega_0}$ and $f=\sinh t$, bounded by graphs $\pm\varphi$ satisfying a first-order equation. Curvature and boundary comparisons force $\Omega_0$ to be a hyperbolic ball and the graph equation makes $\Sigma$ a geodesic sphere, completing the rigidity.

Load-bearing premise

The load-bearing premise is that the flow lines of the normalized gradient of $f$ are geodesics and that the first flow line to leave a boundary component where $f$ is constant meets the other boundary component orthogonally; the warped-product splitting and both rigidity theorems collapse if either geometric assertion fails.

Editorial extensions

If this is right

A direct corollary of Theorem 1.1 is that completeness plus the curvature bounds already imply compactness, with a single boundary component.
In the spherical analogue, the same argument shows that a complete manifold with $\mathrm{Ric}_{\Omega}\ge n$, $H\ge c>0$, and a non-constant solution of $\nabla^2 f + f g=0$ with $f_\nu=c f$ is a geodesic ball of radius $\tan^{-1}(1/c)$ in $\mathbb{S}^{n+1}$.
Under the alternative curvature assumption (K2) with a diameter lower bound of $c/\sqrt{c^2-1}\,\pi$ on boundary components, the boundary must be a round sphere $\mathbb{S}^n$ of radius $c/\sqrt{c^2-1}$, and the manifold is one of two explicit warped products over that sphere.
The Reilly-type corollaries give eigenvalue inequalities in which equality holds exactly for the hyperbolic or spherical geodesic ball, connecting the rigidity to spectral geometry on the boundary.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

The same warped-product and graph-function scheme suggests a quantitative stability version: manifolds that nearly satisfy $\mathrm{Ric}\ge -n$ and $H\ge c$ and admit an approximate solution should lie close to the hyperbolic ball in a Gromov-Hausdorff sense; the paper does not pursue this.
Because the proof separates the graph equation from the curvature comparison, one could test whether $H\ge c$ can be weakened to an integral or average mean-curvature condition while preserving the rigidity, which would link the result to boundary spectral inequalities.
The boundary diameter lower bound in Theorem 1.2 is an input rather than a conclusion; a natural test is whether warped products over compact non-round bases with diameter below that threshold satisfy (K2) and the Robin equation, which would show the bound is necessary rather than technical.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

New result worth knowing: the paper treats the Obata-type equation Hess f = f g with Robin condition f_nu = c f for c>1, a case not covered by earlier work on the + equation or on c<1. Proposition 1.3 gives a warped-product splitting of the manifold without curvature assumptions; that is genuinely new and is the backbone of the paper. The rigidity theorems (1.1, 1.2, 1.4) are plausible and the computations I checked are clean: the derivation of |grad v|=1 and the Gauss equation sum in Proposition 4.2 are correct, and the eigenvalue corollaries are sharp.

The main soft spot is the proof of Theorem 1.1. After deriving Ric_{Omega0} >= -(n-1), H_{partial Omega0} >= c, and the distance bound r <= d(A,partial Omega0), the proof invokes a theorem of Ge (Theorem 0.3 in [6]) to conclude Omega0 is a hyperbolic ball. That theorem is not stated and its hypotheses are not checked: boundary connectedness of partial Omega0, the orientation of the mean-curvature bound, and whether Ge's inequality goes the right way. I could not verify this step from the preprint. It may well be fine -- the inequality direction is consistent with the usual inradius comparison and equality case -- but as written it is a load-bearing black box.

Two smaller things. Proposition 3.1(3) asserts that the exit geodesic meets the other boundary component orthogonally. This is true: at the exit point f equals the extremal value, so the tangential gradient vanishes and grad f is normal. But the paper doesn't say that. And Proposition 2.2 is cited to [2]; it is a one-line computation, so the citation is harmless but the paper could be self-contained.

I also want the authors to clarify how much of Theorems 1.1 and 1.4 overlaps their own [12], since they say they extend Corollaries 4.5 and 4.4 of that paper. Without seeing [12] I can't judge the increment; a referee should ask.

Bottom line: the central argument appears sound, and the gaps are expositional and fixable. I would send this to a serious referee, asking them specifically to verify the application of Ge's theorem and to fill the terse steps. Probably not desk reject.

Referee Report

3 major / 4 minor

Summary. The paper studies the Obata-type equation ∇²f − f g = 0 with Robin boundary condition f_ν = c f, c = coth θ > 1, on a complete connected (n+1)-dimensional manifold with compact boundary. It first proves a structural result (Proposition 1.3) asserting that, depending on whether f is constant on a boundary component, Ω is either a warped product Ω₀ × [−θ,∞) or Ω₀ × [−θ,θ] with metric dt² + cosh²t g_{Ω₀}, or a Z₂-symmetric graph domain over a compact manifold Ω₀ with graph function φ satisfying a first-order PDE. Under the curvature assumption Ric_Ω ≥ −n and boundary mean curvature H ≥ c, Theorem 1.1 concludes that Ω is isometric to a geodesic ball of radius tanh⁻¹(1/c) in hyperbolic space H^{n+1}. Theorem 1.2, under a sectional-curvature type condition (K2) and a lower diameter bound, concludes that boundary components are round spheres and Ω is a warped product over Sⁿ. Theorem 1.4 establishes the analogous spherical rigidity for ∇²f + f g = 0. The paper also derives two eigenvalue inequalities with rigidity statements from Reilly-type formulas.

Significance. If the proof is completed, these results add new cases to the Obata-type rigidity classification for the Robin boundary value problem with c > 1, complementing the c < 1 result of Lai–Zhou and the Euclidean/Neumann results of Xia–Xiong and Chen–Lai–Wang. The paper has several genuine strengths: the ODE derivation of the warped-product metric in Propositions 3.2 and 3.3 is clean; the construction of the distance-like function v with |∇v| ≡ 1 and the distance estimate t ≤ d(A,T_t) are explicit and checkable; the maximum-principle uniqueness of the graph function φ in Proposition 3.9 is sound; and the Gauss-equation computation in Proposition 4.2 is correct. The main theorems are clearly stated and the overall strategy is coherent.

major comments (3)

[§4, proof of Theorem 1.1] The decisive step 'by the fact that tanh⁻¹(1/c) ≤ d(A,∂Ω₀) and Theorem 0.3 in [6], we conclude that Ω₀ is isometric to a geodesic ball ... and A consists of a single point' is not supported as written. Theorem 0.3 of [6] is neither stated nor quoted, and the proof does not verify its hypotheses for the constructed pair (Ω₀, ∂Ω₀): the connectedness of ∂Ω₀ if required, the sign/orientation of the mean-curvature lower bound H_{∂Ω₀} ≥ c, and whether the needed geometric input is a lower bound or an upper bound on the inradius. Since this is the only step that turns Ω₀ into a hyperbolic ball, the authors should state Theorem 0.3 and check each hypothesis explicitly, including the normalization of the hyperbolic space and of the mean curvature.
[§3, Proposition 3.1(3)] The assertion 'We then know that γ'_{p₀}(h(p₀)) ⊥ T_{γ_{p₀}(h(p₀))}Σ' is load-bearing: it is used to derive h(p₀) = 2θ and hence h ≡ 2θ, which in turn yields the warped-product splitting in Proposition 1.3(1). However, no proof or reference is given for this orthogonality. It should be justified by a first-variation argument for the exit time from the boundary component S, or the relevant lemma from [2] should be stated and proved.
[§4, proof of Theorem 1.1, identification of level sets] After applying Theorem 0.3 of [6], the proof says 'we can use the similar method discussed above to prove tanh⁻¹(1/c) − t ≤ d(T_t,∂Ω₀), and we then have T_t = S_t'. This step is necessary for identifying v with the distance from the center and for eventually recognizing Σ as a geodesic sphere through Proposition 2.3, but the argument is omitted. In particular, the text should explain how the two distance lower bounds force equality in the triangle inequality, so that every point of T_t has distance exactly t from the center A = {x₀}.

minor comments (4)

[§2, Proposition 2.2] Proposition 2.2 is stated without proof and cited to [2]; since the fact that the integral curves of ∇f/|∇f| are geodesics is used throughout the paper, a one-line proof (it follows directly from ∇²f = f g together with |∇f|² − f² = 1) or a precise lemma reference in [2] would make the paper more self-contained.
[§4, proof of Theorem 1.1, distance estimate] In the displayed estimate for t ≤ d(A,T_t), the term v(γ(s))|_{l−ε}^{ε} has the integration limits reversed; the correct expression should be v(γ(l−ε)) − v(γ(ε)), or an absolute value should be taken. As written, the displayed equality has the wrong sign.
[Throughout] There are repeated spelling errors: 'isometirc' should be 'isometric' (e.g., in Proposition 1.3, Theorem 1.2, and Proposition 3.2), and 'connetced' should be 'connected' in the proof of Proposition 1.3(2).
[§3, Proposition 3.9] In the uniqueness proof, the monotonicity statement for the function h(t) should explicitly specify the interval (0,θ), and the contradiction should explicitly note that at an interior maximum of φ−ψ one has ∇(φ−ψ)(p) = 0; this is clear but currently implicit.

Circularity Check

0 steps flagged · score 2.0 of 10

No significant circularity: the derivation stands on independent external theorems, with only one non-load-bearing self-citation.

full rationale

The constant c is fixed by the Robin boundary condition in equation (1), not fitted after the fact, and the conclusion that the radius is tanh^{-1}(1/c) is derived from the Obata-type equation rather than assumed. The warped-product splitting in Proposition 1.3 is adapted from the independent source [2], and Proposition 2.2's geodesic claim is cited to [2] as an external lemma; neither is a self-citation. The decisive step in the proof of Theorem 1.1 invokes Theorem 0.3 of [6] after deriving Ric_{Omega_0} >= -(n-1), H_{partial Omega_0} >= c, and tanh^{-1}(1/c) <= d(A, partial Omega_0), and [6] is not authored by the present authors, so this is an external comparison theorem rather than a reduction of the conclusion to its own inputs. The only appearance of the authors' own work is the introductory remark that Theorems 1.1 and 1.4 extend Corollaries 4.5 and 4.4 of [12]; that reference is not cited in either proof and is therefore not load-bearing. The unproved orthogonality assertion in Proposition 3.1(3) and the external geodesic claim in Proposition 2.2 would be correctness gaps if false, but they are not instances of circularity or of fitted inputs being renamed as predictions.

Assumptions & free parameters 0 free parameters · 7 assumptions · 0 invented entities

No free parameters are fitted to data; the constant c is part of the Robin boundary data. The paper relies on standard Riemannian geometry background plus several imported theorems: the geodesic-flow lemma from [2], comparison/rigidity results from [6] (Theorems 0.1 and 0.3) and [10] (Lemma 2.1), the transnormal function result of [17] (Theorem A), and the Reilly-type formula of [14]. These are domain assumptions from the literature, not new postulates.

assumptions (7)

domain assumption Integral curves of grad(f)/|grad(f)| are geodesics (Prop. 2.2)
Stated without proof, cited to [2]; the warped product decomposition in Prop. 1.3 relies on it.
domain assumption Theorem 0.1 in [6]: under Ric >= -n and boundary mean curvature H >= c, the manifold is compact and the boundary is connected.
Used in Theorem 1.1 to reduce to the compact connected-boundary setting so Prop. 1.3(2) applies.
domain assumption Theorem 0.3 in [6]: a manifold with Ric >= -(n-1), boundary mean curvature >= c, and a distance bound from a set A to the boundary yields geodesic-ball rigidity in hyperbolic space.
Used in Theorem 1.1 to conclude Omega_0 is a hyperbolic geodesic ball from the distance inequality tanh^{-1}(1/c) <= d(A, boundary).
domain assumption Lemma 2.1 in [10]: connectivity of the boundary under the same curvature bounds.
Cited in the proof of Theorem 1.1.
domain assumption Theorem 1.3 in [2]: warped product structure for the + Obata equation with Robin boundary condition.
Used in Theorem 1.4 to obtain the Z_2-symmetric warped product representation of Omega.
domain assumption Theorem A in [17] and Lemma 2.6 in [2]: level sets of transnormal functions are submanifolds and the maximum set is connected.
Used in Proposition 4.5 to show the maximum set of f|Sigma is a single point.
standard math Reilly-type formula of [14].
Used in Corollary 4.6 to derive the eigenvalue inequality.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Some rigidity results related to the Obata type equation." pith.science (2026). https://pith.science/paper/JNHTUT6A

@misc{pith2026241118508,
  author       = {Pith},
  title        = {Pith review of: Some rigidity results related to the Obata type equation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/JNHTUT6A}},
  note         = {Machine review of arXiv:2411.18508}
}

abstract

Let $(\Omega^{n+1},g)$ be an $(n + 1)$-dimensional smooth complete connected Riemannian manifold with compact boundary $\partial\Omega=\Sigma$ and $f$ a smooth function on $\Omega$ which satisfies the Obata type equation $\nabla^2 f -fg =0$ with Robin boundary condition $f_{\nu} = cf$, where $c=\coth{\theta}>1$. In this paper, we provide some rigidity results based on the warped product structure of $\Omega$ determined by the equation $\nabla^2 f -fg =0$ and appropriate curvature assumptions. We also apply a similar method to the Obata type equation $\nabla^2 f +fg =0$ and get a rigidity result on the standard sphere $\mathbb{S}^{n+1}$.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

20 extracted references · 20 canonical work pages

[6]

J. Ge. Comparison theorems for manifolds with mean convex boun dary. Commun. Contemp. Math. , 17(5):1550010, 12, 2015

work page 2015
[2]

X. Z. Chen, M. J. Lai, and F. Wang. The Obata equation with Robin boundary condition. Revista matem´ atica iberoamericana, 37(2):643–670, 2020

work page 2020
[12]

Lower bound estimates of the ﬁrst eigen value for boundary of compact manifolds

Yiwei Liu and Yihu Yang. Lower bound estimates of the ﬁrst eigen value for boundary of compact manifolds. Acta Mathematica Sinica, English Series , 2025

work page 2025
[1]

Rigidity on an eigenvalue problem with mixed boundary condition

S. Almaraz and E. Barbosa. Rigidity on an eigenvalue problem with mix ed boundary condition. arXiv preprint arXiv:1710.06701 , 2017

work page Pith review arXiv 2017
[3]

M. P. do Carmo and C. Y. Xia. Rigidity theorems for manifolds with bo und- ary and nonnegative Ricci curvature. Results in Mathematics , 40:122–129, 2001

work page 2001
[4]

J. Escobar. Uniqueness theorems on conformal deformation o f metrics, Sobolev inequalities, and an eigenvalue estimate. Comm. Pure Appl. Math. , 43(7):857–883, 1990

work page 1990
[5]

G. J. Galloway and H. C. Jang. Some scalar curvature warped pro duct splitting theorems. Proc. Amer. Math. Soc. , 148(6):2617–2629, 2020

work page 2020
[7]

Riemannian manifolds with compact boundary

Ryosuke Ichida. Riemannian manifolds with compact boundary. Yokohama Math. J , 29(2):169–177, 1981. REFERENCES 27

work page 1981

Show all 20 references

[8]

M. Kanai. On a diﬀerential equation characterizing a Riemannian st ructure of a manifold. Tokyo J. Math. , 6(1):143–151, 1983

1983
[9]

M. J. Lai and H. H. Zhou. A note on Obata equations on manifolds w ith boundary. J. Math. Study , 55(3):242–253, 2022

2022
[10]

H. Z. Li and Y. Wei. Rigidity theorems for diameter estimates of c ompact manifold with boundary. Int. Math. Res. Not. IMRN , (11):3651–3668, 2015

2015
[11]

Lichnerowicz

A. Lichnerowicz. G´ eom´ etrie des groupes de transformation s. Travaux et Recherches Math´ ematiques III, 1958

1958
[13]

M. Obata. Certain conditions for a Riemannian manifold to be isome tric with a sphere. J. Math. Soc. Japan , 14:333–340, 1962

1962
[14]

G. H. Qiu and C. Xia. A generalization of Reilly’s formula and its applica - tions to a new Heintze–Karcher type inequality. International Mathematics Research Notices, 2015(17):7608–7619, 2015

2015
[15]

Raulot and A

S. Raulot and A. Savo. On the ﬁrst eigenvalue of the Dirichlet-to -Neumann operator on forms. Journal of Functional Analysis , 262(3):889–914, 2012

2012
[16]

R. Reilly. Applications of the Hessian operator in a Riemannian manif old. Indiana University Mathematics Journal , 26(3):459–472, 1977

1977
[17]

Q. M. Wang. Isoparametric functions on Riemannian manifolds. I . Math. Ann., 277(4):639–646, 1987

1987
[18]

G. Q. Wu and R. G. Ye. A note on Obata’s rigidity theorem. Commun. Math. Stat. , 2(3-4):231–252, 2014

2014
[19]

Xia and C

C. Xia and C. W. Xiong. Escobar’s conjecture on a sharp lower bo und for the ﬁrst nonzero Steklov eigenvalue. Peking Mathematical Journal , pages 1–20, 2023

2023
[20]

C. Y. Xia. The ﬁrst nonzero eigenvalue for manifolds with Ricci cu rvature having positive lower bound. Chinese mathematics into the 21st century , 1991. Yiwei Liu: School of Mathematical Sciences, Shanghai Jiao Tong University email: lyw201611012@sjtu.edu.cn Yi-Hu Yang: Schoo...

1991

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.

[1] [6]

J. Ge. Comparison theorems for manifolds with mean convex boun dary. Commun. Contemp. Math. , 17(5):1550010, 12, 2015

work page 2015

[2] [2]

X. Z. Chen, M. J. Lai, and F. Wang. The Obata equation with Robin boundary condition. Revista matem´ atica iberoamericana, 37(2):643–670, 2020

work page 2020

[3] [12]

Lower bound estimates of the ﬁrst eigen value for boundary of compact manifolds

Yiwei Liu and Yihu Yang. Lower bound estimates of the ﬁrst eigen value for boundary of compact manifolds. Acta Mathematica Sinica, English Series , 2025

work page 2025

[4] [1]

Rigidity on an eigenvalue problem with mixed boundary condition

S. Almaraz and E. Barbosa. Rigidity on an eigenvalue problem with mix ed boundary condition. arXiv preprint arXiv:1710.06701 , 2017

work page Pith review arXiv 2017

[5] [3]

M. P. do Carmo and C. Y. Xia. Rigidity theorems for manifolds with bo und- ary and nonnegative Ricci curvature. Results in Mathematics , 40:122–129, 2001

work page 2001

[6] [4]

J. Escobar. Uniqueness theorems on conformal deformation o f metrics, Sobolev inequalities, and an eigenvalue estimate. Comm. Pure Appl. Math. , 43(7):857–883, 1990

work page 1990

[7] [5]

G. J. Galloway and H. C. Jang. Some scalar curvature warped pro duct splitting theorems. Proc. Amer. Math. Soc. , 148(6):2617–2629, 2020

work page 2020

[8] [7]

Riemannian manifolds with compact boundary

Ryosuke Ichida. Riemannian manifolds with compact boundary. Yokohama Math. J , 29(2):169–177, 1981. REFERENCES 27

work page 1981