Combinatorial Calabi flows with ideal circle patterns

Xiaoxiao Zhang

REVIEW 3 major objections 5 minor 30 references

Combinatorial Calabi flows with ideal circle patterns

T0 review · 3 major / 5 minor · reviewed 2026-08-10 · deepseek-v4-flash

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

Pith's one-line read The combinatorial Calabi flow for ideal circle patterns exists for all time and converges exponentially fast to a zero-curvature pattern on a closed surface whenever such a pattern exists.

desk verdict A plausible extension of Calabi flow to ideal circle patterns, but the convergence theorems are not proved: the Euclidean half stops at global existence and the hyperbolic half relies on a missing lemma. read the letter →

arxiv 2501.01605 v1 pith:CZ7TLK54 submitted 2025-01-03 math.DG math.GT

Xiaoxiao Zhang This is my paper · ORCID

classification math.DGmath.GT MSC 53C4452C26

keywords combinatorialCalabiflowidealcirclepatternsdiscretecurvatureenergyhyperbolicbackgroundgeometryEuclideanexponentialconvergencegradient

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

This paper proves that the combinatorial Calabi flow, a discrete analogue of the smooth Calabi flow, can be run on ideal circle patterns in both hyperbolic and Euclidean background geometry. For any initial ideal circle pattern on a closed oriented surface with an edge-weight function satisfying the cell-wise angle sum condition, the flow has a unique solution for all time. If, and only if, an ideal circle pattern with zero curvature exists, the flow converges exponentially fast to that pattern, giving a non-singular hyperbolic metric in the hyperbolic setting and a constant-curvature circle-pattern metric (described in the abstract as a flat cone metric) in the Euclidean setting. This matters because it turns a hard existence question about canonical discrete metrics into a dynamical one: the flow itself finds the pattern, and it does so without the small-energy or extra assumptions needed in earlier hyperbolic Calabi-flow results.

What carries the argument

The engine is the combinatorial Calabi flow $u'(t)=-LK=-\tfrac12\nabla_u C$, the negative gradient flow of the discrete Calabi energy $C(r)=\sum_i K_i^2$; here $u_i=\ln\tanh(r_i/2)$ in hyperbolic geometry and $u_i=\ln r_i$ in Euclidean geometry, and $L$ is the Jacobian of the curvature map $K(u)$. The paper's key structural inputs are the angle-monotonicity and symmetry identities of Lemma 4.2, the positivity of $L$ (Lemma 4.5), a lower bound on a certain area-derivative (Lemma 4.4), and the ideal Ricci potential $F(u)=\int_{\bar u}^{u}\sum_i K_i\,du_i$, whose properness is invoked to force radii to stay away from zero. The edge weights $\Theta$ must satisfy the star condition $\sum_{i=1}^{m}\Theta(e_i)=(m-2)\pi$ around each 2-cell, which is exactly what makes an ideal circle pattern possible.

What would settle it

A direct test is to compute the ideal Ricci potential $F(u)$ for a small explicit ideal circle pattern, such as a tetrahedral pattern on the sphere or a genus-2 triangulation with weights satisfying the star condition, and check whether $F(u)\to+\infty$ as $\|u\|\to+\infty$; a bounded sequence along a ray would disprove the properness used in the hyperbolic proof. A numerical integration of the flow for the same data should also be run: if radii $r_i(t)$ approach zero at finite or infinite time for some initial pattern while a zero-curvature pattern exists, the theorem's conclusion is false.

Watch

Extended reading notes

Core claim

The central claim is stated as Theorems 3.1 and 3.2. Let $D$ be a cellular decomposition of an oriented closed surface, with edge weights $\Theta:E\to(0,\pi)$ satisfying the star condition $\sum_{i=1}^{m}\Theta(e_i)=(m-2)\pi$ on every 2-cell. Under the flow $u'(t)=-LK$, where $u_i=\ln\tanh(r_i/2)$ in the hyperbolic setting, $u_i=\ln r_i$ in the Euclidean setting, $K$ is the discrete curvature vector and $L$ is the Jacobian of $K$ as a function of $u$, any initial ideal circle pattern $r(0)$ produces a global solution. In the hyperbolic case the solution converges exponentially fast to a zero-curvature ideal circle pattern, equivalently to a hyperbolic metric without singularities, if and only if such a pattern exists; in the Euclidean case the analogous statement holds with convergence to a constant-curvature circle-pattern metric if and only if that constant-curvature ideal pattern exists. The proof runs the discrete Calabi energy $C(r)=\|K\|^2$ down its gradient flow, uses positivity of $L$ to get energy decay, and uses compactness of the trajectory plus injectivity of the curvature map to identify the limit.

Load-bearing premise

The load-bearing assumption is that a certain energy function of the circle radii (the ideal Ricci potential) grows without bound as the radii approach zero or infinity; the proof invokes this after equation (4.11) as 'lemma 6.1 in [?]' without stating it, and without that growth the radii could collapse and hyperbolic exponential convergence would fail.

Editorial extensions

If this is right

Any initial ideal Euclidean circle pattern evolves without ever ceasing to exist: radii remain bounded above and the logs of radii grow at most linearly, so the flow is defined for all $t\ge0$.
In the hyperbolic setting, if a zero-curvature ideal circle pattern exists, it is unique, and every trajectory converges to it with exponential decay of the discrete Calabi energy $C(r(t))$.
The 'only if' direction gives an obstruction: if no zero-curvature ideal circle pattern exists, the hyperbolic Calabi flow cannot converge, even though the solution still exists for all time.
Since convergence is characterized by existence of a target pattern, the flow can be used constructively: numerically following $u'=-LK$ is a method for finding zero-curvature ideal circle patterns on a given surface.
The Euclidean convergence statement parallels the hyperbolic one with the target replaced by a constant-curvature ideal circle-pattern metric, so the two background geometries are unified by the same gradient-flow mechanism.

Reading between the lines

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

Editorial inference: the hyperbolic half of the theorem inherits its guarantee that radii do not collapse from a properness claim about the ideal Ricci potential that the manuscript invokes after equation (4.11) as 'lemma 6.1 in [?]' without stating or referencing it; if that properness fails, the convergence proof would need a different lower-bound argument.
Editorial inference: the same gradient-flow mechanism—positive definiteness of the curvature Jacobian plus properness of the associated potential—should transfer to other discrete curvature functionals, so the paper suggests a general template for discrete Calabi-type flows.
Editorial inference: the Euclidean bound on $|\omega_{ij}|\le c(\Theta)$ yields only exponential-in-time bounds on radii; a sharper estimate using dissipation of the Calabi energy might upgrade global existence to a bound independent of time, which would be a natural test of the method.
Editorial inference: the combinatorial conditions from the Ricci-flow theorems (H3–H5 and E3–E4) characterize existence of zero-curvature patterns; combined with this paper's Calabi-flow convergence, they give a purely combinatorial criterion for when the discrete Calabi flow lands on a canonical metric.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

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

Bottom line: the paper is not close to proving what it claims. It extends combinatorial Calabi flow to ideal circle patterns, a natural next step after Ge-Hua-Zhou's Ricci flow and Ge's Calabi flow for ordinary circle patterns, and the flow definition is sensible. But the abstract promises exponential convergence in both geometries, and neither theorem delivers it.

Start with the Euclidean part, because it is cleanly checkable. Section 5 proves only global existence: the author bounds |du_i/dt| by a constant, integrates to get two-sided exponential bounds on r_i(t), and stops. There is no Calabi-energy decay estimate, no compactness of the trajectory, no use of the positive-definite Jacobian to drive K to a constant vector, and no exponential rate. The convergence and iff claims of Theorem 3.2 are never addressed. The stress-test note is right: this is not a missing detail, it is the main statement.

The hyperbolic proof has a different, equally serious hole. In Section 4, after constructing the ideal Ricci potential F, the lower bound on the radii r_i(t) depends on 'By lemma 6.1 in [?], lim F(u) = +∞' (just after (4.11)). That lemma is not in the bibliography. Without it, compactness of the trajectory in R_{>0}^{|V|} fails, and the exponential convergence argument for Theorem 3.1 collapses. The 'only if' direction of Theorem 3.1 is fine, and the rest of the 'if' direction would be standard if the properness were supplied.

Credit where it is due. The paper is the first to write down this flow for ideal circle patterns, and some of the technical pieces are correct: Lemma 4.4, the uniform bound on |ω_ij| in Section 5, and the use of Ge-Hua-Zhou's rigidity results. The acknowledgment of a competing independent result (Shengyu Li and Zhigang Wang) is honest. The setup and notation are clear.

Who is this for? A specialist in discrete conformal geometry might profit from the formulation and might be able to repair the gaps by adapting Ge-Xu or Ge-Hua's Calabi-flow arguments. But as submitted, the main theorems are unsupported. I would not cite it as a convergence theorem. If it is submitted to a journal, the right response is a reject/resubmit, not a full referee cycle: ask the author to supply the missing properness lemma and to actually prove the Euclidean convergence. A serious editor should not send this to referees in its current state.

Referee Report

3 major / 5 minor

Summary. The manuscript defines combinatorial Calabi flows for ideal circle patterns in hyperbolic and Euclidean background geometry (Definition 2.1 and equation (3.5), with the Euclidean analogue in Section 5) and claims, in Theorems 3.1 and 3.2, global existence and exponential convergence to a nonsingular hyperbolic metric or to a constant-curvature flat cone metric, respectively, with an if-and-only-if characterization in terms of the existence of a zero-curvature or constant-curvature ideal circle pattern. The hyperbolic proof uses the ideal Ricci potential and known rigidity results of Ge-Hua-Zhou; the Euclidean proof derives a uniform bound on the logarithmic radii and concludes global existence.

Significance. If the results were correctly proved, they would be a natural and useful extension of Ge-Hua-Zhou's combinatorial Ricci flow theorems for ideal circle patterns to Calabi-type flows, and the use of discrete Calabi energy, positive definiteness of the curvature Jacobian, and Andreev-Thurston rigidity is a reasonable strategy. The manuscript also helpfully recalls the relevant equivalence conditions H1-H5 and E1-E4 from Ge-Hua-Zhou. However, as written, the main convergence and equivalence claims are not established: the Euclidean proof stops after global existence, and the hyperbolic proof depends on an uncited properness lemma. The significance of the paper is therefore conditional on substantial additional arguments.

major comments (3)

[Section 5, proof of Theorem 3.2] The proof of Theorem 3.2 derives only global existence. After bounding |ω_ij| and hence |du_i/dt|, the text concludes with the estimate c0 e^{-c1 t} ≤ r_i(t) ≤ c0 e^{c1 t} and states that the flow has a solution for all time. Yet Theorem 3.2 also asserts exponential convergence to a constant-curvature circle pattern metric and an equivalence between convergence and the existence of such a pattern. No Lyapunov or Calabi-energy monotonicity argument is given, no compactness of the trajectory in R_{>0}^{|V|} is established, no use is made of the positive definiteness of the Euclidean Jacobian L, and no exponential-rate estimate appears. The Euclidean half of the abstract and Theorem 3.2 is therefore unsupported by the proof as written.
[Section 4, after equation (4.11)] The proof of the 'if' direction of Theorem 3.1 relies on the assertion 'By lemma 6.1 in [?], there holds lim_{||u||→+∞, u∈R^{|V|}_{<0}} F(u)=+∞.' This properness of the ideal Ricci potential is load-bearing: it is used to obtain the uniform lower bound r_i(t) ≥ C > 0, which prevents radii from collapsing and is then used for compactness and exponential convergence. The lemma is neither stated nor cited with a valid reference. Without a proof or a correct citation, the hyperbolic convergence theorem is not established. The missing reference must be supplied and the proof of the properness step must be included.
[Section 4, proof of Theorem 3.1, '⇒' direction] The argument 'u(n+1)-u(n) = u'(ξ_n)' applies the scalar mean value theorem to a vector-valued function without justification; one would need to apply the mean value theorem componentwise or use an integral identity. This is not the central obstruction, but it is part of the convergence argument and should be fixed.

minor comments (5)

[Section 4, proof of Lemma 4.4] The sentence 'By lemma 4.3, we know ∂θ_j/∂r_i > 0' is a mis-citation: Lemma 4.3 states that θ_i becomes small when r_i is large, while the positivity of ∂θ_j/∂r_i follows from Lemma 4.2(ii) together with the identity ∂θ_i/∂r_j sinh r_j = ∂θ_j/∂r_i sinh r_i. Please correct the citation.
[Throughout] The placeholder citation 'By lemma 6.1 in [?]' must be replaced with a real statement and reference; the same applies to any other unresolved '?' markers.
[Section 5] In the final line of the proof of Theorem 3.2, 'for all time y ∈ [0, ∞)' should read 't ∈ [0, ∞)'.
[Theorems 3.1-3.2] The phrase 'constant curvature circle pattern metric' in Theorem 3.2 and 'flat cone metric' in the abstract is not defined precisely; in particular, it should be stated whether 'constant curvature' means all vertex curvatures equal to a common value, and how that value is determined.
[Section 2.1] The definition of an ideal circle pattern refers to circles meeting at an 'interior common point'; the distinction between interior and exterior common points is supported only by figures and would benefit from a precise geometric definition in the text.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the derivation is not self-referential, and the identified gaps are omissions rather than circular reductions.

full rationale

The paper's main results (Theorems 3.1 and 3.2) are derived from external results attributed to other authors: Ge-Hua-Zhou [16] for curvature-map injectivity, positive definiteness of the Jacobian L, and existence/rigidity of ideal circle patterns; Andreev-Thurston rigidity; and Bobenko-Springborn/Rivin for existence of ideal D-type circle patterns under condition (⋆). None of these are the present author's own prior results, so there is no self-citation chain that forces the conclusions. The combinatorial Calabi flow is defined independently as the negative gradient flow with respect to u = ln tanh(r/2) or u = ln r, and no fitted parameter is later renamed as a prediction. The convergence argument in Theorem 3.1 uses standard Lyapunov/compactness reasoning: F decreases, properness is invoked to prevent radii collapsing, L positive definiteness gives exponential decay of the Calabi energy. Even though the properness assertion 'By lemma 6.1 in [?], lim F(u)=+∞' appears with an undefined reference, that is a missing proof or missing citation, not a circular reduction: the statement is not shown to be equivalent to the theorem being proved. Similarly, the Euclidean proof in Section 5 explicitly derives only global existence (the bounds |du_i/dt| ≤ c1 and r_i(t) bounded by exponentials), while the abstract and Theorem 3.2 additionally assert exponential convergence and an 'if and only if' characterization; those parts are not supplied in the text. This is a serious incompleteness of the proof, but it is not circularity in the sense of the target conclusion being built into an input or a fitted parameter. No step of the derivation reduces by definition to the theorem's conclusion, and no load-bearing claim is justified solely by a self-citation. The correct finding is therefore 'no significant circularity,' with a score of 0.

Assumptions & free parameters 0 free parameters · 5 assumptions · 0 invented entities

No free parameters are fitted or chosen by hand; the edge weights Theta are fixed input data satisfying condition (*). No new physical or geometric entities are introduced. The main external support comes from the rigidity and positive-definiteness results of Ge-Hua-Zhou, plus a properness lemma that is cited with a missing reference.

assumptions (5)

domain assumption The curvature map K for ideal circle patterns is injective (rigidity).
Used in the proof of Theorem 3.1 to identify the unique zero-curvature limit pattern; cited to Ge-Hua-Zhou [16] and built on Andreev-Thurston and Bobenko-Springborn.
domain assumption The Jacobian matrix L = dK/du is positive definite along the flow.
Lemma 4.5 quoted from [16]; it drives energy decay and yields K=0 from LK=0 in the hyperbolic proof.
ad hoc to paper The ideal Ricci potential F(u) is proper on R^|V|_<0, so F(u) tends to +infinity as ||u|| tends to infinity.
Invoked as 'lemma 6.1 in [?]' with no bibliographic entry; it provides the uniform lower bound on radii needed for compactness and convergence.
standard math For any two radii and intersection angle in (0, pi), a unique Euclidean or hyperbolic two-circle configuration exists.
Lemma 4.1; this underlies the definition of the piecewise flat or hyperbolic cone metric associated to a radius vector.
domain assumption For any weight satisfying condition (*), an ideal D-type circle pattern exists on surfaces of any genus.
Justifies the ideal circle pattern setting; credited to Rivin [25] for genus zero and Bobenko-Springborn [26] for higher genus.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Combinatorial Calabi flows with ideal circle patterns." pith.science (2026). https://pith.science/paper/CZ7TLK54

@misc{pith2026250101605,
  author       = {Pith},
  title        = {Pith review of: Combinatorial Calabi flows with ideal circle patterns},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/CZ7TLK54}},
  note         = {Machine review of arXiv:2501.01605}
}

read the original abstract

In this paper, we extend the work of Ge-Hua-Zhou \cite{GHZ} on combinatorial Ricci flows for ideal circle patterns to combinatorial Calabi flows in both hyperbolic and Euclidean background geometry. We prove the solution to the combinatorial Calabi flows with any given initial Euclidean (hyperbolic resp.)ideal circle pattern exists for all time and converges exponentially fast to a flat cone metric (hyperbolic resp.) on a given surface.

Figures

Figures reproduced from arXiv: 2501.01605 by the authors.

**Figure 3.** pattern of two circles [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

30 extracted references · 29 canonical work pages

[1]

E. M. Andreev, On convex polyhedra of finite volume in Loba ˘cevki˘i spaces, Math. USSR-Sb., 12(1970), 255-259

work page 1970
[2]

E. M. Andreev, On convex polyhedra in Loba˘ cevki˘i spaces, Math. USSR-Sb., 10(1970), 412- 440. 14

work page 1970
[3]

X. Bao, F. Bonahon, Hyperideal polyhedra in hyperbolic 3-space , Bull. Soc. Math. France, 130(2002), 457-491

work page 2002
[4]

Bobenko, B.A

A.I. Bobenko, B.A. Springborn, Variational principles for circle patterns and Koebe¡ ¯s theo- rem,Trans. Am. Math. Soc. 356 (2004), 659 ¨C689

work page 2004
[5]

Chow, The Ricci flow on the 2-sphere , J

B. Chow, The Ricci flow on the 2-sphere , J. Differential Geom, 33(2)(1991), 325-334

work page 1991
[6]

Chow and F

B. Chow and F. Luo, Combinatorial Ricci flows on surfaces , J. Differ. Geom., 63 (2003), 97-129

work page 2003
[7]

Chang, The 2-dimensional Calabi flow , Nagoya Math

S. Chang, The 2-dimensional Calabi flow , Nagoya Math. J. 181(2006), 63-73. MR2210710

work page 2006
[8]

Chang, Global existence and convergence of solutions of Calabi flow on surfaces of genus , J

S. Chang, Global existence and convergence of solutions of Calabi flow on surfaces of genus , J. Math. Kyoto Univ. 40(2)(2000), 363-377, DOI 10.1215/kjm/1250517718

work page arXiv 2000

Show all 30 references

[9]

Chen, Calabi flow in Riemann surfaces revisited: a new point of view , Internat

X. Chen, Calabi flow in Riemann surfaces revisited: a new point of view , Internat. Math. Res. Notices 6(2001), 275-297, DOI 10.1155/S1073792801000149

2001 doi
[10]

Calabi, Extremal K¨ ahler metrics, Seminar on Differential Geometry, Ann

E. Calabi, Extremal K¨ ahler metrics, Seminar on Differential Geometry, Ann. Of Math. stud., vol. 102, Princeton Univ. Press, princeton Univ. Press, Princeton, N.J., 1982, 259-190

1982
[11]

Calabi, Extremal K¨ ahler metrics.II , Differential geometry and complex analysis, Springer, Berlin, 1985, 95-114

E. Calabi, Extremal K¨ ahler metrics.II , Differential geometry and complex analysis, Springer, Berlin, 1985, 95-114

1985
[12]

R. S. Hamilton, The Ricci flow on surfaces, Math. General Relativity (Santa Cruz, CA, 1986), 237¨C262, Contemp. Math., 71, Amer. Math. Soc., Providence, RI, 1988

1986
[13]

Ge, Combinatorial Calabi flows on surfaces , Trans

H. Ge, Combinatorial Calabi flows on surfaces , Trans. Amer. Math. Soc, 370(2)(2018), 1377- 1391

2018
[14]

Ge, Combinatorial methods and geometric equations , Thesis (Ph

H. Ge, Combinatorial methods and geometric equations , Thesis (Ph. D.)-Peking University, Beijing. 2012. 144pp

2012
[15]

Ge and B

H. Ge and B. Hua, On combinatorial Calabi flows with hyperbolic circle patterns , Advances in Mathematics, 333(2018),523-538

2018
[16]

H. Ge, B. Hua and Z. Zhou, Combinatorial Ricci flows for ideal circle patterns , Advances in Mathematics, 383(2021), 107698

2021
[17]

Ge and W

H. Ge and W. Jiang, On the deformation of inversive distance circle packings , II. J. Funct. Anal. 272(9)(2017), 3573-3595

2017
[18]

Ge and X

H. Ge and X. Xu, A discrete Ricci flow on surfaces with hyperbolic background geometry , Int. Math. Res. Not. IMRN. (11)(2017), 3510-3527

2017
[19]

Ge and X

H. Ge and X. Xu, 2-dimensional combinatorial Calabi flow in hyperbolic background geometry, Differ. Geom. Appl., 47(2016), 86-98

2016
[20]

Guo, Local rigidity of inversive distance circle packing , Trans

R. Guo, Local rigidity of inversive distance circle packing , Trans. Amer. Math. Soc. 363 (2011),4757-4776

2011
[21]

Guo and F

R. Guo and F. Luo, Rigidity of polyhedral surfaces, II, Geom. Topol. 13 (2009), 1265-1312

2009
[22]

Koebe, Kontaktprobleme der konformen Abbildung Ber

P. Koebe, Kontaktprobleme der konformen Abbildung Ber. S¨achs. Akad. Wiss. Leipz., Math.- Phys. Kl., 88(1936), 141-164

1936
[23]

Marden and B

A. Marden and B. Rodin, On Thurston¡ ¯s formulation and proof of Andreev¡ ¯s theorem , Computational methods and function theory (Valparaso, 1989), 103-115, Lecture Notes in Math., 1435, Springer, Berlin, 1990. 15

1989
[24]

Rousset, Sur la rigidit´ e de poly` edres hyperboliques en dimension 3: cas de volume fini, cas hyperid´ eal, cas fuchsien, Bull

M. Rousset, Sur la rigidit´ e de poly` edres hyperboliques en dimension 3: cas de volume fini, cas hyperid´ eal, cas fuchsien, Bull. Soc. Math. France. 132(2004), 233-261

2004
[25]

Rivin, A characterization of ideal polyhedra in hyperbolic 3-space, Ann

I. Rivin, A characterization of ideal polyhedra in hyperbolic 3-space, Ann. of Math. 143 (1996), 51-70

1996
[26]

B. A. Springborn, A variational principle for weighted Delaunay triangulations and hyperideal polyhedra, J. Differential Geom. 78(2008), 333-367

2008
[27]

J. M. Schlenker, Hyperideal circle patterns, Math. Res. Lett. 12(1)(2005), 85-102

2005
[28]

Thurston, Geometry and topology of 3-manifolds , Princeton lecture notes 1976

W. Thurston, Geometry and topology of 3-manifolds , Princeton lecture notes 1976

1976
[29]

Y. C. de Verdi` ere, Un principe variationnel pour les empilements de cercles , Invent. Math. 104(3)(1991), 655-669

1991
[30]

Zhou, Circle patterns, topological degrees and deformation theory , arXiv:1703.01768

Z. Zhou, Circle patterns, topological degrees and deformation theory , arXiv:1703.01768. 16

arXiv

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.

[1] [1]

E. M. Andreev, On convex polyhedra of finite volume in Loba ˘cevki˘i spaces, Math. USSR-Sb., 12(1970), 255-259

work page 1970

[2] [2]

E. M. Andreev, On convex polyhedra in Loba˘ cevki˘i spaces, Math. USSR-Sb., 10(1970), 412- 440. 14

work page 1970

[3] [3]

X. Bao, F. Bonahon, Hyperideal polyhedra in hyperbolic 3-space , Bull. Soc. Math. France, 130(2002), 457-491

work page 2002

[4] [4]

Bobenko, B.A

A.I. Bobenko, B.A. Springborn, Variational principles for circle patterns and Koebe¡ ¯s theo- rem,Trans. Am. Math. Soc. 356 (2004), 659 ¨C689

work page 2004

[5] [5]

Chow, The Ricci flow on the 2-sphere , J

B. Chow, The Ricci flow on the 2-sphere , J. Differential Geom, 33(2)(1991), 325-334

work page 1991

[6] [6]

Chow and F

B. Chow and F. Luo, Combinatorial Ricci flows on surfaces , J. Differ. Geom., 63 (2003), 97-129

work page 2003

[7] [7]

Chang, The 2-dimensional Calabi flow , Nagoya Math

S. Chang, The 2-dimensional Calabi flow , Nagoya Math. J. 181(2006), 63-73. MR2210710

work page 2006

[8] [8]

Chang, Global existence and convergence of solutions of Calabi flow on surfaces of genus , J

S. Chang, Global existence and convergence of solutions of Calabi flow on surfaces of genus , J. Math. Kyoto Univ. 40(2)(2000), 363-377, DOI 10.1215/kjm/1250517718

work page arXiv 2000