Pith. sign in

REVIEW 3 major objections 5 minor 37 references

The mean curvature flow of subgroups on Lie groups of dimension three

T0 review · 3 major / 5 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read Every 2D subgroup of a non-unimodular 3D Lie group with non-vanishing mean curvature evolves under mean curvature flow by an explicit eternal solution.

desk verdict The subgroup classification and translator theorem are solid, but the paper's advertised non-translating eternal solutions rest on a sign error: the only fully worked verification solves the reverse flow, so Theorem 1.2 is not supported as stated. read the letter →

arxiv 2505.17892 v2 pith:ZPSHN6GP submitted 2025-05-23 math.DG

Romina M. Arroyo , Gabriela P. Ovando , Mariel Sáez This is my paper · ORCID
classification math.DG MSC 53A1053C4253C4422E2522F30
keywords meancurvatureflowLiesubgroupsleft-invariantmetricstranslatorssolitonssolvmanifoldsnon-unimodulargroupsthree-dimensional
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 mean curvature flow of two-dimensional Lie subgroups inside three-dimensional Lie groups equipped with left-invariant metrics. It establishes that in the unimodular case every Lie subgroup is minimal, hence a static solution, while in the non-unimodular case every abelian subgroup with non-vanishing mean curvature evolves by translations and every non-abelian subgroup with non-vanishing mean curvature evolves by explicit families written in Theorem 3.3, all existing for all time with constant mean curvature and without self-similarity. A four-dimensional example exhibits a non-abelian nilpotent subgroup that evolves by translation, supporting the conjecture that in higher dimensions the translating class is the nilpotent one.

What carries the argument

The central identity is Lemma 2.1: for a subgroup $K$ with unit normal $\nu$, the scalar mean curvature is $H=-\operatorname{tr}(\operatorname{ad}\nu)$, so $H$ is constant along $K$ and vanishes precisely for unimodular groups. Around this, the classification of two-dimensional subgroups of the solvable families $S_{3,\lambda}$, $S_{3,\lambda_1,\lambda_2}$, $S'_{3,\lambda}$, and of $\mathrm{SL}(2,\mathbb{R})$ supplies the candidate initial data; translator solutions are obtained from the Killing-field condition $g(V,\nu)=-H$, and non-translating solutions are obtained from the ansatz that replaces the subgroup parameter $A$ in the factor $Ae^{\lambda z}$ by $Ac(t)e^{\lambda z}$, which turns the flow into an ODE for $c(t)$.

What would settle it

For the unverified family $\varphi_2(y,z,t)=\left(\frac{e^{\lambda_1 z+(\lambda_1+\lambda_2)\lambda_1 t}-1}{\lambda_1}a,\,y,\,z\right)$ on $S_{3,\lambda_1,\lambda_2}$ with $\lambda_1\neq\lambda_2,\lambda_2\neq0$, directly compute the induced metric, unit normal, and mean curvature $H_t$ of the surface at a fixed time $t>0$, and compare $-H_t$ with $g(\partial_t\varphi_2,\nu_t)$. If they do not match identically, the listed solution is not a mean curvature flow and Theorem 1.2 fails as stated.

Watch

Extended reading notes

Core claim

For a non-unimodular three-dimensional Lie group with the left-invariant metric (4), every two-dimensional Lie subgroup with non-vanishing mean curvature yields an explicit solution of the mean curvature flow. The abelian subgroups are translators along a Killing direction (Theorem 3.2), and the non-abelian subgroups evolve by the one-parameter families listed in Theorem 3.3, each existing for all time with constant mean curvature and without self-similarity. The mechanism is algebraic: by Lemma 2.1 the scalar mean curvature of a subgroup equals $-\operatorname{tr}(\operatorname{ad}\nu)$ for a unit normal $\nu$, so it is constant; the translators are found from $g(V,\nu)=-H$, and the non-translating evolutions come from the ansatz that promotes the subgroup parameter $A$ in the factor $Ae^{\lambda z}$ to $Ac(t)e^{\lambda z}$, reducing the flow to an ordinary differential equation for $c(t)$. The paper also gives a four-dimensional example in which a Heisenberg subgroup, non-abelian and nilpotent, evolves by translation.

Load-bearing premise

The non-translating evolutions for all but one of the non-abelian families in Theorem 3.3 are asserted by analogy: the paper verifies the computation only for the family $K_{1,b}$ in $S_{3,\lambda_1,\lambda_2}$, and if the analogous computation fails for any of the other listed families, Theorem 1.2 overstates what is proven.

Editorial extensions

If this is right

  • On non-unimodular three-dimensional Lie groups, every abelian two-dimensional subgroup with non-vanishing mean curvature is a translator, so its mean curvature flow is a one-parameter family of isometries that exists for all time.
  • The non-abelian evolutions listed in Theorem 3.3 are explicit eternal solutions with constant mean curvature at each time, and they are not self-similar, giving concrete non-trivial examples of mean curvature flow in a curved ambient space.
  • In three dimensions, abelian subgroups are exactly the translating ones; the four-dimensional example shows the class widens to nilpotent subgroups in higher dimensions, as the paper conjectures.
  • The explicit parametrizations, with all Lie-group and subgroup parameters kept, provide the basis for the paper's announced study of the moduli space of solutions as the parameters vary.

Reading between the lines

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

  • If the analogous computations for the remaining families check out, the paper effectively gives a complete description of the mean-curvature flow of every codimension-one Lie subgroup in a three-dimensional solvable Lie group with the orthonormal left-invariant metric (4), not just existence of some solution.
  • The ansatz $Ac(t)e^{\lambda z}$ suggests that for solvable groups with diagonal dilations the flow acts by rescaling the subgroup parameter; a natural test is to replace the scalar $c(t)$ by a matrix $C(t)$ when multiple dilation factors are present, as in $S'_{3,\lambda}$.
  • The four-dimensional translator example suggests that in higher-dimensional solvable Lie groups the class of subgroups that evolve by translations may be exactly the nilpotent ones, which would give a clean algebraic characterization of self-similar translating solutions.
  • Since the non-translating solutions retain a constant mean curvature in space at each time, they are candidates for long-time limits or singularity models in which curvature does not blow up but the shape changes through the time-dependent parameter only.
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.

Referee Report

3 major / 5 minor

Summary. The paper studies the mean curvature flow of two-dimensional Lie subgroups inside three-dimensional Lie groups equipped with left-invariant metrics. It classifies all such subgroups, shows that in unimodular groups every Lie subgroup is minimal (hence static), and then for non-unimodular groups constructs two families of explicit solutions: translating solutions (for abelian subgroups and some exceptional non-abelian cases) and non-translating solutions for non-abelian subgroups with non-vanishing mean curvature. The main theorems (Theorems 1.1, 1.2, and 3.3) assert that these evolutions exist for all time, have constant mean curvature at each time, and that the non-abelian evolutions are not self-similar. A four-dimensional example of a non-abelian translating subgroup is also given.

Significance. If the results were correct, the paper would provide a complete catalogue of explicit eternal solutions to the mean curvature flow with constant mean curvature in non-Euclidean three-dimensional manifolds, which would be a valuable contribution to the theory of geometric flows in homogeneous spaces. The translator computations in Section 3, based on Lemma 2.1 and explicit Killing fields, are concrete and largely verifiable, and the idea of introducing a time-dependent ansatz for non-translating evolutions is promising. However, the sign error in the only fully worked verification (Section 4.2.1) and the misclassification of a family of subgroups in Proposition 2.3 mean that the main theorems are not supported as stated. The paper does not provide machine-checked proofs or reproducible code; its strengths are the explicit algebraic computations and the clear structural approach, but these are undermined by the load-bearing errors detailed below.

major comments (3)
  1. [§4.2.1 and Theorem 3.3] The verification of the evolution of K_{1,b} in S_{3,λ1,λ2} contains a sign error. For A = b e^{(λ1+λ2)λ2 t}, the unit normal is ν = (A E3 − E2)/√(1+A^2) and Lemma 2.1 gives H = −(λ1+λ2)A/√(1+A^2). Yet ∂φ/∂t = (λ1+λ2)A E2, so g(∂φ/∂t, ν) = −(λ1+λ2)A/√(1+A^2) = H, not −H. Thus the listed φ1 does not satisfy equation (1); it satisfies the reverse flow. The same sign issue affects the S_{3,λ} solution in Theorem 3.3, where with A = a e^{λ^2 t} one obtains g(∂φ/∂t, ν) = −λ A/√(1+A^2) while −H = 2λ A/√(1+A^2). Since the remaining families are stated to be analogous and left to the reader, Theorem 1.2 and the non-translating parts of Theorem 3.3 are not supported as written.
  2. [Proposition 2.3(iii)(a) and Theorem 3.2] The subgroup K_{1,b} = {(x, bz, z)} for λ2 = 0 is misclassified as abelian. Its Lie subalgebra is spanned by E1 and E3 + bE2, and [E3 + bE2, E1] = λ1 E1 ≠ 0, so it is non-abelian for λ1 ≠ 0. Consequently Theorem 3.2's assertion that a subgroup solves (3) if and only if it is abelian or has H = 0 is false: for b ≠ 0, K_{1,b} is non-abelian, H = −λ1 b/√(1+b^2) ≠ 0, and Section 3.2.2 itself shows it is a translator. This also contradicts the abstract and introduction's claim that non-abelian evolutions are never self-similar.
  3. [§4.2 and Theorem 3.3] The paper states that 'the other ones are analogous' and leaves the remaining non-translating evolutions to the reader. Because the single detailed verification in §4.2.1 is incorrect, the claim that the listed families in Theorem 3.3 solve (1) is unsupported. At minimum, the correct ODE for c(t) must be derived and verified for every family (with the correct sign in the time exponent) before the eternal-solution claim can be accepted.
minor comments (5)
  1. [§4.2.1] In the definition of ν^t, the denominator appears as 1 + b^2 e^{-2(λ1+λ2)λ2 t}, but the later computations and consistency with A = b e^{(λ1+λ2)λ2 t} require 1 + b^2 e^{2(λ1+λ2)λ2 t}.
  2. [§4.1] For S_{3,λ} the translating solution is written as φ(x,z,t) = (x, y, 2λt), but y is not a variable of φ; the intended expression is φ(x,y,t) = (x, y, 2λt).
  3. [§5] In the displayed metric, the term '− x/2 e^{-2s}dydz' appears; it should presumably involve dxdz, since a dydz term would duplicate the preceding term. Please check the metric and the subsequent normal computation.
  4. [Abstract and Introduction] The statement that evolutions are self-similar for abelian subgroups but not self-similar in the other cases is inconsistent with the translator solution for K_{1,b} in the λ2 = 0 case once the classification is corrected; the summary should be revised accordingly.
  5. [Section 3.2.2] The phrase 'the Lie subgroup K_{2,a} with Lie subalgebra k_{1,b} = span{E2, E3 + aE1}' contains an indexing error: the subalgebra should be denoted k_{2,a}, not k_{1,b}.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: the central derivation is self-contained, with only minor overlapping-author citations and an explicitly flagged completeness gap in the non-translating verification.

full rationale

The paper's core derivation is not circular. Mean curvature of a Lie subgroup is computed from first principles as H = -tr(ad nu) (Lemma 2.1, proved via the Koszul formula), and the self-similar condition g(V,nu) = -H is solved explicitly family by family in Section 3, using Killing vector bases that are either computed from left translations or, for the S_{3,lambda,lambda} hyperbolic case, taken from the overlapping-author reference [8]. The non-translating Theorem 3.3 solutions are introduced as an explicit ansatz (replacing A e^{lambda z} by A c(t) e^{lambda z}, Section 4.2) and then checked against the flow equation; the only fully worked check in Section 4.2.1 computes the mean curvature using connection formulas attributed to [8, Sections 3.2 and 5.1], but those formulas are standard and independently derivable from the paper's own Koszul setup, so they do not import the target result. No parameter is fitted to data, and no 'prediction' is defined in terms of the quantity it claims to derive. The paper itself flags a completeness limitation: Section 4.2 states that 'the other ones are analogous, are left to the interested reader', so the proof of Theorem 3.3 relies on unshown computations for all families except K_{1,b} in S_{3,lambda_1,lambda_2}; this is a proof gap and a possible correctness risk, but not circularity. The overlap with [8] is a minor self-citation and is not load-bearing in the sense of reducing the argument to its own claim. Hence the circularity score is low.

Assumptions & free parameters 0 free parameters · 4 assumptions · 0 invented entities

No free parameters were fitted; the constants λ, λ1, λ2 and subgroup parameters a,b,c,d,e,f are classification data, not ad hoc choices. No new entities are postulated. The main unexamined inputs are the non-compact MCF uniqueness assumption and the deferred verification of the ansatz.

assumptions (4)
  • standard math Classification of three-dimensional real Lie algebras and their two-dimensional subalgebras as listed in Section 2.1.
    Invoked throughout Section 2 to enumerate the ambient groups and subgroups; cited to [33] and [19].
  • standard math Ferrand-Obata theorem on inessential conformal transformations: on manifolds not conformal to S^n or R^n, conformal Killing fields with C≠0 do not exist, so self-similar solutions reduce to Killing (translating) flows.
    Used in Section 1 to justify restricting to Killing vector fields; theorem proved in [3,4,17,18].
  • domain assumption Short-time existence and uniqueness of mean curvature flow for codimension-one hypersurfaces in Riemannian manifolds, which guarantees that the explicit one-parameter families solving (1) are the evolution of the initial subgroup.
    Assumed implicitly when the paper calls the explicit families 'the evolution' of a subgroup; standard for smooth compact hypersurfaces but not stated for these noncompact families.
  • standard math Connection and mean-curvature formulas from [8, Sections 3.2 and 5.1] used in the detailed verification in §4.2.1.
    These formulas are derived from the Koszul formula and are standard; the overlap of authors with [8] does not make the target result circular.

how reviews work

0 comments
Cite this review

Pith. "Pith review of The mean curvature flow of subgroups on Lie groups of dimension three." pith.science (2026). https://pith.science/paper/ZPSHN6GP

@misc{pith2026250517892,
  author       = {Pith},
  title        = {Pith review of: The mean curvature flow of subgroups on Lie groups of dimension three},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/ZPSHN6GP}},
  note         = {Machine review of arXiv:2505.17892}
}
read the original abstract

In this work we study the existence of solutions to the Mean Curvature Flow for which the initial condition has the structure of a two-dimensional Lie subgroup within a Lie group of dimension three. We consider Lie groups with a fixed left-invariant metric and first observe that if the Lie group is unimodular, then every Lie subgroup is a minimal surface (hence a trivial solution). For this reason we focus on non-unimodular Lie groups, finding the evolution of every Lie subgroup of dimension 2 (within a 3 dimensional Lie group). These evolutions are self-similar for abelian subgroups (i.e. evolve by isometries), but not self-similar in the other cases.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

37 extracted references · 36 canonical work pages

  1. [1]

    Andrews and X

    B. Andrews and X. Z. Chen , Curvature flow in hyperbolic spaces , J. Reine Angew. Math. 729 (2017), 29--49; MR3680369

    work page 2017

  2. [2]

    Abresch and J

    U. Abresch and J. Langer , The normalized curve shortening flow and homothetic solutions , J. Differential Geom., 23 (2), 175--196 (1986)

    work page 1986

  3. [3]

    Alekseevsky , Groups of conformal transformations of Riemannian spaces , Math

    D.V. Alekseevsky , Groups of conformal transformations of Riemannian spaces , Math. Sb., 89 (1), 280--296 (1972)

    work page 1972

  4. [4]

    Alekseevsky , The sphere and the Euclidean space are the only Riemannian manifolds with essential conformal transformations , Uspekhi Math

    D.V. Alekseevsky , The sphere and the Euclidean space are the only Riemannian manifolds with essential conformal transformations , Uspekhi Math. Nauk, 28 (5), 289--290 (1973)

    work page 1973

  5. [5]

    Anastassiou, I

    S. Anastassiou, I. Chrysikos , The R icci flow approach to homogeneous Einstein metrics on flag manifolds , J. Geom. Phys. 61 (2011), no. 8, 1587–1600

    work page 2011

  6. [6]

    Andrews, B Chow, C

    B. Andrews, B Chow, C. Guenther and M. Langford , Extrinsic Geometric Flows . Graduate Studies in Mathematics, (2020)

    work page 2020

  7. [7]

    Arroyo, G

    R. Arroyo, G. Ovando, R. Perales, M. S\'aez , The mean curvature flow on solvmanifolds , Bol. Soc. Mat. Mex. 30 (2), Art. 39 (2024)

    work page 2024

  8. [8]

    B\"ohm , On the long time behavior of homogeneous R icci flows , Comment

    C. B\"ohm , On the long time behavior of homogeneous R icci flows , Comment. Math. Helv. 90 (2015), no. 3, 543–571

    work page 2015

Show all 37 references
  1. [9]

    B\"ohm, R

    C. B\"ohm, R. A. Lafuente , Immortal homogeneous R icci flows , Invent. Math. 212 (2018), no. 2, 461–529

    2018

  2. [10]

    B\"ohm, R

    C. B\"ohm, R. A. Lafuente , The Ricci flow on solvmanifolds of real type , Adv. Math. vol. 352 (2019), 516-540

    2019

  3. [11]

    Bourni, M

    T. Bourni, M. Langford and G. Tinaglia , Translating solutions to mean curvature flow , Minimal surfaces: integrable systems and visualisation, Springer Proc. Math. Stat., 349 , 1--12 (2021). https://doi.org/10.1007/978-3-030-68541-6_1

    2021 doi

  4. [12]

    Ortiz , Rotational surfaces with prescribed mean curvature in H^2

    A.Bueno, I. Ortiz , Rotational surfaces with prescribed mean curvature in H^2 . Ann. Mat. Pura Appl. 4 201 (3), 1257–1277 (2022)

    2022

  5. [13]

    Ortiz , A new family of translating solitons in hyperbolic space

    A.Bueno, I. Ortiz , A new family of translating solitons in hyperbolic space . Preprint. arXiv:2402.05533

    arXiv

  6. [14]

    K. Choi, R. Haslhofer and O. Hershkovits , Enhanced profile estimates for ovals and translators , Adv. Math. 453 (2024), Paper No. 109853, 22 pp.; MR4779319

    2024

  7. [15]

    Cosgaya and S

    A. Cosgaya and S. Reggini , Isometry groups of three-dimensional Lie groups , Ann. Global Anal. Geom. 61 (4), 831--845 (2022)

    2022

  8. [16]

    Ferrand J , The action of conformal transformations on a Riemannian manifold , Math

    J. Ferrand J , The action of conformal transformations on a Riemannian manifold , Math. Ann. 304 (2), 277--291 (1996)

    1996

  9. [17]

    Ferrer, F

    L. Ferrer, F. Martín, R. Mazzeo, M. Rodríguez , Properly embedded minimal annuli in H^2 R . Math. Ann. 375 , 541--594 (2019). https://doi.org/10.1007/s00208-019-01840-5

    2019 doi

  10. [18]

    Frances , Sur les vari\'et\'es lorentziennes dont le groupe conforme est essentiel , Math

    C. Frances , Sur les vari\'et\'es lorentziennes dont le groupe conforme est essentiel , Math. Ann. 332 (1), 103--119 (2005)

    2005

  11. [19]

    Gorbatsevich, A

    V. Gorbatsevich, A. Onishchik, E. Vinberg , Structure of Lie groups and Lie algebras . English transl. in Encycl. Math Sc. 41 , Springer-Verlag, Berlin, Heidelberg, 1994

    1994

  12. [20]

    K. Y. Ha and J. B. Lee , The isometry groups of simply connected 3 -dimensional unimodular Lie groups . J. Geom. Phys. 62 , 189--203 (2012)

    2012

  13. [21]

    Hoffman, T

    D. Hoffman, T. Ilmanen, F. Mart\' n, and B. White , Minimal surfaces: integrable systems and visualisation, 147--168, Springer Proc. Math. Stat., 349, Springer, Cham, (2021) https://doi.org/10.1007/978-3-030-68541-6_9

    2021 doi

  14. [22]

    Hoffman, F

    D. Hoffman, F. Mart\' n, and B. White Scherk-like translators for mean curvature flow. J. Differential Geom. 122 (3), 421--465 (2022)

    2022

  15. [23]

    Huisken , Contracting convex hypersurfaces in Riemannian manifolds by their mean curvature , Invent

    G. Huisken , Contracting convex hypersurfaces in Riemannian manifolds by their mean curvature , Invent. math. 84 , 463--480 (1986)

    1986

  16. [24]

    Huisken and A

    G. Huisken and A. Polden , Geometric evolution equations for hypersurfaces . Calculus of variations and geometric evolution problems (Cetraro, 1996), 45--84, Lecture Notes in Math. 1713 , Fond. CIME/CIME Found. Subser., Springer, Berlin, (1999)

    1999

  17. [25]

    Isenberg, M

    J. Isenberg, M. Jackson , Ricci flow of locally homogeneous geometries on closed manifolds , J. Differential Geom. 35 (1992), no. 3, 723–741

    1992

  18. [26]

    Isenberg, M

    J. Isenberg, M. Jackson, P. Lu , Ricci flow on locally homogeneous closed 4- manifold , Comm. Anal. Geom. 14 (2006), no. 2, 345–386

    2006

  19. [27]

    Lauret , The Ricci flow for simply connected nilmanifolds , Comm

    J. Lauret , The Ricci flow for simply connected nilmanifolds , Comm. Anal. Geom. 19 (2011), 831–854

    2011

  20. [28]

    Lauret , Ricci flow of homogeneous manifolds , Math

    J. Lauret , Ricci flow of homogeneous manifolds , Math. Z. 274 (2013), 373–403

    2013

  21. [29]

    J. H. S. de Lira and F. Mart\'in Serrano , Translating solitons in Riemannian products, J. Differential Equations 266 (2019), no. 12, 7780--7812; MR3944241

    2019

  22. [30]

    Lott , On the long time behaviour of type-III R icci flow solutions , Math

    J. Lott , On the long time behaviour of type-III R icci flow solutions , Math. Annalen (2007), no. 3, 627–666

    2007

  23. [31]

    W. H. Meeks III, J. P\'erez , Constant mean curvature surfaces in metric Lie groups , Contemp. Math. 570 (2011), 25--110

    2011

  24. [32]

    Colombo, L

    G. Colombo, L. Mari, and M. Rigoli , Remarks on mean curvature flow solitons in warped products . Discrete Contin. Dyn. Syst. Ser. S. 13 (7), (2020) 1957--1991

    2020

  25. [33]

    Milnor , Curvatures of Left Invariant Metrics on Lie Groups , Adv

    J. Milnor , Curvatures of Left Invariant Metrics on Lie Groups , Adv. in Math. 21 , 293--329 (1976)

    1976

  26. [34]

    Obata, The conjectures of conformal transformations of Riemannian manifolds , Bull

    M. Obata, The conjectures of conformal transformations of Riemannian manifolds , Bull. Amer. Math. Soc. 77 (1971), 265--270; MR0270397

    1971

  27. [35]

    T. L. Payne , The R icci flow for nilmanifolds , J. Mod. Dyn. 4 (2010), 65–90

    2010

  28. [36]

    Pipoli , Invariant translators of the solvable group , Annali di Matematica Pura ed Applicata 199 , 1961--1978 (2020)

    G. Pipoli , Invariant translators of the solvable group , Annali di Matematica Pura ed Applicata 199 , 1961--1978 (2020)

    2020

  29. [37]

    Pipoli , Invariant Translators of the Heisenberg Group , J

    G. Pipoli , Invariant Translators of the Heisenberg Group , J. of Geom. Analysis 31 , 5219--5258 (2021)

    2021

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.