On the long-time behavior of immortal Ricci flows

Shaosai Huang

REVIEW 3 major objections 5 minor 49 references

On the long-time behavior of immortal Ricci flows

T0 review · 3 major / 5 minor · reviewed 2026-08-14 · deepseek-v4-flash

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

Pith's one-line read Immortal Ricci flows with bounded curvature and diameter sub-converge to compact Riemannian orbifolds, and type-III flows with $t\mu_+'(t)\to 0$ are non-collapsing with negative Einstein blowdown limits.

desk verdict Serious preprint; Theorem 1.1's proof hinges on the author's own unpublished heat-density estimates, and the negative-Einstein claim in Theorem 1.3 is cited en bloc rather than proven. read the letter →

arxiv 1908.05410 v1 pith:R4A6WGW2 submitted 2019-08-15 math.DG

Shaosai Huang This is my paper · ORCID

classification math.DG MSC 53E2053C23

keywords immortalRicciflowtype-IIIcollapsinggeometryorbifoldlimitsentropyfunctionalsnegativeEinsteinmanifoldnon-collapsingGromov-Hausdorffconvergence

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 asks what happens to an immortal Ricci flow—a solution defined for all positive time on a closed manifold—as time goes to infinity, under the uniform curvature-diameter bound $\operatorname{diam}(M,g(t))^2\sup_M|\operatorname{Rm}|\le D^2$. It establishes that any unbounded sequence of time slices sub-converges, in the Gromov-Hausdorff sense, to a compact Riemannian orbifold: volume collapsing can occur, but only down to orbifold singularities, never to the corner singularities that appear in arbitrary collapsing sequences. For type-III flows whose diameter grows at most like $t^{1/2}$, it proves a second result: if the derivative of the $\mu_+$-entropy satisfies $\limsup_{t\to\infty} t\mu_+'(t)=0$, the global volume ratio cannot degenerate, and every blowdown limit is an $m$-dimensional negative Einstein manifold. A sympathetic reader would care because these are structural statements about long-time limits of immortal Ricci flows that do not assume static or soliton data from the start.

What carries the argument

The load-bearing object is the unwrapped neighborhood $W_{x_0}$ around an orbifold point $x_0$ of the collapsing limit. Passing to the fiberwise universal cover removes the nilpotent lattice actions, gives a uniform injectivity radius, and produces a limit metric $\tilde g_\infty$ on $W_{x_0}$ together with a Riemannian submersion $p:(W_{x_0},\tilde g_\infty)\to(V_{x_0},\hat g_X)$ whose fibers are simply connected nilpotent Lie groups; the commuting family of Killing fields tangent to the fibers, the limit central distribution $\mathcal C$, has volume density $\sqrt{\det H}$ that locally represents $\chi_C$. The asymptotic vanishing of the functional derivatives yields the gradient steady or expanding soliton equation for $\tilde g_\infty$, and the submersion curvature identities convert it into an elliptic equation for $\log\det H$ whose structure supports the maximum-principle argument: the maximum of $\chi_C$ lies in the orbifold part, and the equation forces $\chi_C$ to be constant, eliminating corner singularities.

What would settle it

Construct an immortal Ricci flow on a closed $m$-manifold ($m\ge3$) satisfying the uniform curvature-diameter bound, with bounded curvature and diameter, such that some sequence $(M,g(t_i))$ converges in the Gromov-Hausdorff sense to a space containing a corner singularity—for instance a limit interval where the generic fiber is a 2-torus and over some points the fiber is a circle; such a limit would refute Theorem 1.1. Alternatively, to test Theorem 1.3, compute $t\mu_+'(t)$ along a type-III flow with $\operatorname{diam}(M,g(t))=O(t^{1/2})$ whose global volume ratio tends to zero; if $t\mu_+'(t)\to0$, the non-collapsing conclusion is contradicted.

Watch

Extended reading notes

Core claim

The central discovery is that Ricci flow evolution itself is strong enough to rule out the worst collapsing singularity type. When a sequence of time slices $(M,g(t_i))$ collapses to a lower-dimensional limit $(X,d)$, the a priori limit is an orbifold with corners; the corner set $\tilde S$ is characterized as the zero locus of a limit central density $\chi_C$. The paper shows that along an immortal Ricci flow the asymptotic vanishing of the $F$-entropy derivative forces the limit metric on unwrapped neighborhoods to satisfy the gradient steady soliton equation, hence a submersion-curvature elliptic equation for $\log\det H$ with nonnegative right-hand side; the maximum principle then forces $\chi_C$ to be constant, so $\tilde S$ is empty. In the type-III case, the analogous use of the $W_+$-entropy and the hypothesis $t\mu_+'(t)\to0$ produces a gradient expanding soliton equation with an extra positive term $\dim Z$; the maximum principle forces $\dim Z=0$, contradicting collapse of the global volume ratio and yielding the non-collapsing conclusion.

Load-bearing premise

The argument depends on uniform bounds for the volume-normalized conjugate heat densities $\rho_i=u(t_i)$ on the collapsing time slices—$|M|_{g(t_i)}u(t_i)$ between $C^{-1}$ and $C$ and $|M|_{g(t_i)}|\nabla\rho_i|\le C$—estimates imported from the paper's own unpublished companion preprint [31, Props. 5.1 and 5.3]; if those fail, the limit integrand $w_X$ need not vanish and the gradient soliton equation on the unwrapped neighborhoods does not follow.

Editorial extensions

If this is right

Under the assumptions of Theorem 1.1, every unbounded sequence of time slices has a subsequence converging to a compact Riemannian orbifold, so the long-time limit singular set can only consist of orbifold points.
If a time slice is sufficiently collapsed along such a flow, the underlying manifold is an infranil fiber bundle over a compact lower-dimensional orbifold, equivalently a manifold carrying a pure polarized $F$-structure.
For type-III flows with diameter growth $O(t^{1/2})$ and $t\mu_+'(t)\to0$, the global volume ratio $|M|_{g(t)}\operatorname{diam}(M,g(t))^{-m}$ has a positive lower bound, so the flow does not asymptotically collapse.
Any blowdown limit of such a type-III flow is an $m$-dimensional negative Einstein manifold.
The theorem gives a Ricci-flow counterpart of a classical volume non-collapsing theorem for negatively Ricci-curved manifolds with bounded curvature and diameter.

Reading between the lines

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

If the same maximum-principle and elliptic-equation mechanism extends to other geometric flows possessing monotone entropy functionals, one would expect analogous orbifold-only compactness for their long-time limits.
A testable weakening of Theorem 1.3 would replace $\limsup t\mu_+'(t)=0$ by an integrability condition such as $\int^\infty t\mu_+'(t)\,dt<\infty$; the proof only needs a sequence of times along which the limit integrand vanishes, so the dichotomy may persist under milder decay.
In dimension three, combining this orbifold-limit statement with known long-time behavior could further restrict which geometric model pieces an immortal Ricci flow can collapse to, with orbifold points corresponding to quotients of the pieces.
The local tools (unwrapped neighborhoods, integral convergence, central density) are formulated for arbitrary collapsing sequences with bounded curvature and diameter, so they may be useful for measured Gromov-Hausdorff limits of other families with no Ricci flow structure present.

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

This preprint is worth your time, but read it with the dependencies in mind. The two main theorems are genuinely new statements. Theorem 1.1 — immortal Ricci flow with uniformly bounded curvature and diameter subconverges to a compact Riemannian orbifold — is a strong structural result, and Theorem 1.3 gives a μ_+-based entropy criterion forcing non-collapsing and negative-Einstein blowdowns for type-III flows. If correct, both are significant. The architecture is coherent: collapse → unwrap → integral convergence → limit soliton → maximum principle → orbifold; the F/μ_+ derivative vanishing is the engine, and Sections 3–5 build a local toolbox (Theorems 1.4–1.6) that is largely proven in the text and should be reusable.

Now the soft spots, proportionally. The main one is Proposition 6.2, which for Theorem 1.1 imports uniform control of the conjugate heat densities ρ_i = u(t_i) — specifically |M| ρ_i ∈ [C^{-1}, C] and |M| |∇ρ_i| ≤ C — from the same author's arXiv preprint [31, Props 5.1, 5.3] under only the collapsed regime bound (1.2). This is exactly the setting with no injectivity radius and no volume-ratio lower bound. If those estimates fail or require extra hypotheses, w_X ≡ 0 does not follow, (6.2) has no basis, and Theorem 1.1's corner-singularity exclusion collapses. It is not an obvious error, but a referee must verify [31] or require a self-contained proof. Theorem 1.3's proof, by contrast, is mostly self-contained: Proposition 6.3 derives the analogous bounds from the Euler-Lagrange equation for the μ_+ minimizer. The dependency is concentrated on Theorem 1.1.

Two smaller items. First, the abstract claims the blowdown limit is negative Einstein, but the proof of Theorem 1.3 only establishes non-collapsing; the negative-Einstein step is asserted via an en-bloc citation to [10, 26, 41, 16], with no named theorem or argument. Second, the second fundamental form bound (5.5) for torus orbits is sketched in a paragraph, not proven; likely fixable, but it is a gap.

Who should read it: people working in collapsing geometry, long-time Ricci flow, and entropy rigidity. It deserves peer review, not a desk reject. If I were an editor I'd send it to a strong geometer and ask them to focus on the [31] estimates and on pinning down the expander rigidity citation.

Recommendation: engage with it, conditionally. If the dependencies check out, this is a solid, useful paper.

Referee Report

3 major / 5 minor

Summary. The paper studies long-time limits of immortal Ricci flows on closed m-dimensional manifolds (m >= 3) satisfying the uniform curvature-diameter bound (1.2). Theorem 1.1 asserts that if curvature and diameter stay uniformly bounded in time, then every unbounded sequence of time slices has a subsequence converging to a compact Riemannian orbifold. Theorem 1.3 asserts that for a type-III flow with diam(M, g(t)) = O(t^(1/2)), the condition lim sup_{t→∞} t mu_+'(t) = 0 forces a uniform positive lower bound on the global volume ratio, i.e. non-collapsing; the abstract and introduction add that any blowdown limit is an m-dimensional negative Einstein manifold. The proofs introduce local tools: unwrapped neighborhoods around orbifold points (Theorem 1.4), a collapsing-and-integral-convergence theorem (Theorem 1.5), and a limit central density whose zero locus is the corner singular set (Theorem 1.6). The main results are then derived by showing that the limit metric on the unwrapped neighborhoods satisfies a gradient steady or expanding Ricci soliton equation, and by applying a maximum principle to the central density to exclude corner singularities.

Significance. If correct, the paper would establish a strong structural rigidity statement: long-time limits of controlled immortal Ricci flows are orbifolds, and the entropy condition prevents volume collapsing in a natural type-III class. The local structural theorems (1.4 through 1.6) are deliberately formulated without the global frame-bundle construction, so they may be reusable in other collapsing problems; the short proof of Rong's theorem in Section 7 is a nice illustration. The paper is honest about which parts are standard and which parts are imported, and the overall proof strategy is coherent. However, the main results currently rest on estimates imported from an unpublished preprint, the proof of one advertised conclusion (negative Einstein blowdown) is not actually supplied, and the construction of the central density contains a key estimate that is only sketched. These issues must be resolved before the central claims are fully supported.

major comments (3)

[Section 6.1, Proposition 6.2] The proof of Proposition 6.2 obtains the uniform bounds sup |ln(rho_i |M|)| <= C and sup |grad rho_i| |M| <= C from [31, Props 5.1 and 5.3] and [47, Thm 3.3], but those propositions are not stated, [31] is the author's unpublished preprint, and the setting is exactly the collapsing regime, where no injectivity radius or volume-ratio lower bound is available. If the quoted estimates require any additional hypothesis beyond (1.2), then the equality w_X = 0 in the proof of Proposition 6.2 does not follow, equation (6.2) (the limit gradient steady soliton equation) is unsupported, and the maximum-principle step in Section 6.2 that rules out corner singularities has no basis. This is load-bearing for Theorem 1.1.
[Sections 1 and 6.2] The abstract and the introduction (page 3) assert that under lim sup_{t→∞} t mu_+'(t) = 0 any blowdown limit is an m-dimensional negative Einstein manifold, citing [10], [26], [41] and [16] en bloc. However, Theorem 1.3 as proved in Section 6.2 only establishes the non-collapsing inequality (1.3); after the contradiction argument the text does not derive the additional soliton rigidity that would identify the blowdown limit as Einstein. Either supply a precise statement and proof of the rigidity step with a named theorem, or weaken the advertised conclusion to the non-collapsing statement (1.3).
[Section 5.1, inequality (5.5)] The uniform second fundamental form bound sup |II_{T_i}(z)| <= C_{5.5} is asserted via a sketch combining (5.3) with the flatness of the connection along the T_i orbits, but the actual estimate is not written out and the dependence on the regularity constants is not tracked. This bound is needed to control the sectional curvature of (F M_i/T_i, [\bar g^1_i]) and hence to define the positive continuous density \bar chi_Q that underpins Theorem 1.6 and the subsequent maximum-principle argument. Please provide a complete proof with explicit constants.

minor comments (5)

[Throughout] There are many typos and misspellings, e.g. 'relavent', 'correpsonding', 'ﬁberation', 'comapct', 'Thoerem', and 'functioanl'; the manuscript needs a careful proofreading pass.
[Figures and diagrams] The commutative diagrams (2.8), (3.8) and (5.8) are not legible in the current PDF because of the ASCII-style formatting; they should be typeset properly.
[Introduction, after (1.6)] The phrase 'ln(det G)−1/2 u∞-harmonic function' is confusing; please clarify the operator and the meaning of this terminology.
[Section 2.3 and references] Since [31] supplies central estimates for Proposition 6.2, please state the exact hypotheses and results used from [31] (or include them in an appendix), so that the reader can verify that they apply under only the assumptions of Theorem 1.1.
[Section 4.1, proof of Proposition 4.1] The notation for the averaged density is inconsistent: after introducing rho_{X,i}(x), the proof later refers to 'rho_{X,i}' without the argument in the estimates following (4.2), and the notational dependence on i should be made explicit throughout.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the theorems are derived from stated hypotheses; the self-cited preprint [31] supplies independent analytic estimates rather than the target conclusions.

full rationale

The derivation chain for Theorems 1.1 and 1.3 does not reduce to its hypotheses. Theorem 1.3 uses the vanishing of t mu_+'(t) as an input and derives non-collapsing; the conclusion is not used to establish the entropy condition. Proposition 6.3 verifies the needed density bounds internally from the mu_+ minimizer via the Euler-Lagrange equation, Cheng-Yau estimates, and a Harnack inequality, so Theorem 1.3 is genuinely derived. The main possible concern is Proposition 6.2, where the uniform C^1 bounds on the conjugate heat densities rho_i are imported from the author's own unpublished preprint [31] together with [47]. This is a load-bearing citation, but it is not circular: [31] is cited for independent analytic estimates (heat kernel magnitude and gradient bounds under curvature-diameter control), not for the orbifold or non-collapsing conclusions of this paper, and no fitted parameter is renamed as a prediction. Whether those estimates are valid under exactly the hypotheses (1.2) is a verification or correctness question, not an instance of the paper's conclusion being assumed. The en bloc citation of [10], [26], [41], [16] for the Einstein property of blowdowns similarly represents an unstated known-result dependency rather than a circular step. Accordingly the appropriate circularity score is 0.

Assumptions & free parameters 0 free parameters · 6 assumptions · 0 invented entities

The paper introduces no new physical or extra-mathematical entities. The 'limit central distribution' and 'limit central density' chi_C are proof-internal constructions built from the limit of the centers of the nilpotent groups; they are well-defined mathematical objects constructed from the data, not postulated independently. The ledger above records the prior results and external estimates the central claim rests on that are not proved inside the paper.

assumptions (6)

standard math Cheeger-Fukaya-Gromov nilpotent structure and invariant metrics [11, Thm 2.6, Prop 4.3]; Fukaya's singular fibrations [19, Thm 0.12], [20, Thm 0-7].
The entire collapsing framework in Sections 2.1 through 3 is built on these prior theorems, used as black boxes for the existence of the infranil fiber bundle and the approximating invariant metrics.
standard math Unwrapped neighborhoods have a uniform injectivity radius lower bound, cited from [38, Lemma 2.5] and asserted in Section 3.3.1.
Used to take C^infinity limits of the covering metrics; if false, the limit soliton metric on W_{x0} need not exist.
standard math Shi's derivative estimates [46] give uniform C^l curvature bounds (1.4) for time slices of Ricci flows with bounded curvature.
Invoked at the start of Section 1 and used throughout to apply the collapsing machinery to time slices of the flow.
standard math Feldman-Ilmanen-Ni theory of W_+ and mu_+ [16]: monotonicity, differentiability, minimizer existence and smoothness, and formulas (2.17) and (2.18).
Basis for Proposition 6.3 and for the entropy-growth criterion in Theorem 1.3.
domain assumption Heat kernel density estimates: |M| u(t) and |M| |grad u(t)| uniformly bounded along immortal flows with bounded curvature and diameter, from Huang [31, Props 5.1 and 5.3] and Zhang [47, Thm 3.3].
Load-bearing for Proposition 6.2: these bounds are needed to apply Proposition 4.2 with rho_i = u(t_i). [31] is an unpublished preprint by the same author, so this carries a self-citation burden.
standard math Mal'cev rigidity and affine classification of the fibers [11, Thm 3.7], and the uniform gap for finite subgroups of GL(k0, Z) [6].
Used in Section 5.3 to control the fundamental group action on the central fibers and to define the constant epsilon_0.

how reviews work

0 comments

Cite this review

Pith. "Pith review of On the long-time behavior of immortal Ricci flows." pith.science (2026). https://pith.science/paper/R4A6WGW2

@misc{pith2026190805410,
  author       = {Pith},
  title        = {Pith review of: On the long-time behavior of immortal Ricci flows},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/R4A6WGW2}},
  note         = {Machine review of arXiv:1908.05410}
}

abstract

For an immortal Ricci flow on an $m$-dimensional $(m\ge 3)$ closed manifold, we show the following convergence results: (1) if the curvature and diameter are uniformly bounded, then any unbounded sequence of time slices sub-converges to a Riemannian orbifold; (2) if the flow is type-III with diameter growth controlled by $t^{\frac{1}{2}}$, then any blowdown limit is an $m$-dimensional negative Einstein manifold, provided that Feldman-Ilmanen-Ni's $\boldsymbol{\mu}_+$-functional satisfies $\lim_{t\to \infty} t\boldsymbol{\mu}_+'(t)=0$.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

49 extracted references · 49 canonical work pages

[31]

Improving Subseasonal Forecasting in the Western U.S. with Machine Learning

Shaosai Huang, Notes on Ricci ﬂows with collapsing init ial data (I): Distance distortion. Preprint, arXiv: 1809.07394

work page Pith review arXiv
[10]

Je ﬀCheeger, Mikhail Gromov and Michael Taylor, Finite propaga tion speed, kernel estimates for functions of the Laplace operator, and the geometry of complete Riemannian m anifolds. J. Di ﬀerential Geom. 17 (1982), no. 1, 15-53

work page 1982
[26]

Richard Hamilton, A compactness property for solution s of the Ricci ﬂow. Amer . J. Math.117 (1995), 545-572

work page 1995
[41]

Preprint, arXiv: math / 0211159

Grisha Perelman, The entropy formula for the Ricci ﬂow a nd its geometric applications. Preprint, arXiv: math / 0211159

work page
[16]

Mikhail Feldman, Tom Ilmanen and Lei Ni, Entropy and red uced distance for Ricci expanders. J. Geom. Anal. 15 (2005), no. 1, 49-62

work page 2005
[1]

Bamler, Long-time behavior of 3-dimensional Ricci ﬂow: introduction

Richard H. Bamler, Long-time behavior of 3-dimensional Ricci ﬂow: introduction. Geom. Topol. 22 (2018) 757-774

work page 2018
[2]

Besse, Einstein manifolds

Arthur L. Besse, Einstein manifolds. Ergebnisse der Mat hematik und ihrer Grenzgebiete (3), Springer-V erlag, Berlin,

work page
[3]

Christoph Böhm, On the long time behavior of homogeneous Ricci ﬂows. Comment. Math. Helv. 90 (2015), no. 3, 543-571

work page 2015

Show all 49 references

[4]

Lafuente and Miles Simon, Opti mal curvature estimates for homogeneous Ricci ﬂows

Christoph Böhm, Ramiro A. Lafuente and Miles Simon, Opti mal curvature estimates for homogeneous Ricci ﬂows. Int. Math. Res. Not. rnx256, https://doi.org/10.1093/imrn/rnx256

doi
[5]

Lafuente, Immortal homogen eous Ricci ﬂows

Christoph Böhm and Ramiro A. Lafuente, Immortal homogen eous Ricci ﬂows. Invent. Math. 212 (2018), no. 2, 461-529

2018
[6]

Astérisque, 81

Peter Buser and Hermann Karcher, Gromov’s almost ﬂat man ifolds. Astérisque, 81. Société Mathématique de France, Paris, 1981. 148 pp

1981
[7]

Je ﬀCheeger and Mikhail Gromov, Collapsing Riemannian manifol ds while keeping their curvature bounded. I. J. Diﬀerential Geom. 23 (1986), no. 3, 309-346

1986
[8]

Je ﬀCheeger and Mikhail Gromov, Collapsing Riemannian manifol ds while keeping their curvature bounded. II. J. Diﬀerential Geom. 32 (1990), no. 1, 269-298

1990
[9]

Jianguo Cao, Je ﬀCheeger and Xiaochun Rong, Splitting and Cr-structures for manifolds with nonpositive sectional curvature. Invent. Math. 144 (2001), no. 1, 139-167

2001
[11]

Je ﬀCheeger, Kenji Fukaya and Mikhail Gromov, Nilpotent struct ures and invariant metrics on collapsed manifolds. J. Amer . Math. Soc.5 (1992), 327-372

1992
[12]

Je ﬀCheeger and Xiaochun Rong, Existence of polarized F-structures on collapsed manifolds with bounded curvature and diameter. Geom. Funct. Anal. 6 (1996), no. 3, 411-429

1996
[13]

Shiu Y uen Cheng and Shing-Tung Yau, Di ﬀerential equations on Riemannian manifolds and their geome tric appli- cations. Comm. Pure Appl. Math. 28(1975), 333-354

1975
[14]

Manuscripta Math

Xianzhe Dai, Guofang Wei and Rugang Ye, Smoothing Riema nnian metrics with Ricci curvature bounds. Manuscripta Math. 90 (1996), 49-61

1996
[15]

Matias del Hoyo and Rui Loja Fernandes, Riemannian metr ics on Lie groupoids. J. Reine Angew. Math. 735 (2018), 143-173

2018
[17]

Kenji Fukaya, Collapsing of Riemannian manifolds and e igenvalues of Laplace operator. Invent. Math. 87 (1987), 517-547

1987
[18]

Kenji Fukaya, Collapsing Riemannian manifolds to ones of lower dimensions. J. Di ﬀerential Geom. 25 (1987), 139-156

1987
[19]

Kenji Fukaya, A boundary of the set of the Riemannian man ifolds with bounded curvatures and diameters. J. Diﬀer- ential Geom. 28 (1988), 1-21

1988
[20]

Kenji Fukaya, Collapsing Riemannian manifolds to ones with lower dimension II. J. Math. Soc. Japan 41 (1989), 333-356

1989
[21]

Eduardo Gallego, Luciano Gualandri, Gilbert Hector an d Agustí Reventós, Groupoïdes Riemanniens. Publ. Mat. 33 (1989), No. 3, 417-422

1989
[22]

Alexander Gorokhovsky and John Lott, The index of a tran sverse Dirac-type operator: the case of abelian Molino sheaf. J. Reine. Angew. Math. 678 (2013), 125-162

2013
[23]

Mikhail Gromov, Manifolds with negative curvature. J. Diﬀerential Geom. 13 (1978), no. 2, 223-230

1978
[24]

Mikhail Gromov, Almost ﬂat manifolds. J. Diﬀerential Geom. 13 (1978), no. 2, 231-241

1978
[25]

44 SHAOSAI HUANG

Mikhail Gromov, Structures métriques pour les variété s Riemanniennes, Editions Cedic, Paris (1981). 44 SHAOSAI HUANG

1981
[27]

Surveys in Diﬀ

Richard Hamilton, Formation of singularities in the Ri cci ﬂow. Surveys in Diﬀ. Geom. 2 (1995), 7-136

1995
[28]

Richard Hamilton, Non-singular solutions of the Ricci ﬂow on threemanifolds. Comm. Anal. Geom. 7 (1999), 695- 729

1999
[29]

Preprint, arXiv: 1411.6058

Christian Hilaire, Ricci ﬂow on Riemannian groupoids. Preprint, arXiv: 1411.6058

arXiv
[30]

Shaosai Huang, ε-Regularity and structure of four dimensional shrinking Ri cci solitons. Int. Math. Res. Not. rny069, https://doi.org/10.1093/imrn/rny069

doi
[32]

John Lott, Some geometric properties of the Bakry-Émer y-Ricci tensor. Comment. Math. Helv. 78 (2003), 865-883

2003
[33]

John Lott, On the long-time behavior of type-III Ricci ﬂ ow solutions, Math. Ann. 339 (2007), 627-666

2007
[34]

John Lott, Dimensional reduction and the long-time beh avior of Ricci ﬂow, Comment. Math. Helv. 85 (2010), 485- 534

2010
[35]

To appear, Comment

John Lott, The collapsing geometry of Ricci-ﬂat 4-mani folds. To appear, Comment. Math. Helv., arXiv: 2017.06780

2017
[36]

John Milnor, Curvature of left invariant metrics on Lie groups. Adv. Math. 21 (1976), 293-329

1976
[37]

Translated fro m the French by Grant Cairns

Pierre Molino, Riemannian foliations. Translated fro m the French by Grant Cairns. With appendices by G. Cairns, Y . Carrière, É. Ghys, E. Salem and V . Sergiescu. Progress in Mathematics, 73. Birkhäuser Boston, Inc., Boston, MA ,
[38]

Surveys in geometric analysis and relativity, 439-466, Adv

Aaron Naber and Gang Tian, Geometric structures of coll apsing Riemannian manifolds I. Surveys in geometric analysis and relativity, 439-466, Adv. Lect. Math. (ALM), 2 0, Int. Press, Somerville, MA , 2011

2011
[39]

Aaron Naber and Gang Tian, Geometric structures of coll apsing Riemannian manifolds II. J. Reine Angew. Math. 744 (2018), 103-132

2018
[40]

Michigan Math

Barrett O’Neill, The fundamental equations of a submer sion. Michigan Math. J. 13 (1966), 459-469

1966
[42]

Preprint, arXiv: math /0303109

Grisha Perelman, Ricci ﬂow with surgery on three-manif olds. Preprint, arXiv: math /0303109
[43]

Duke Math

Xiaochun Rong, A Bochner theorem and applications. Duke Math. J. 91 (1998), no. 2, 381-392

1998
[44]

Rothaus, Logarithmic Sobolev inequalities an d the spectrum of Schrödinger operators J

Oscar S. Rothaus, Logarithmic Sobolev inequalities an d the spectrum of Schrödinger operators J. Funct. Anal. 42 (1981), 110-120

1981
[45]

Ruh, Almost ﬂat manifolds

Ernst A. Ruh, Almost ﬂat manifolds. J. Diﬀerential Geom. 17 (1982), 1-14

1982
[46]

Wan-Xiong Shi, Deforming the metric on complete Rieman nian manifolds. J. Diﬀerential Geom. 30 (1989), no. 1, 223-301

1989
[47]

Zhang, Some gradient estimates for the heat kernel equation on domains and for an equation by Perelman

Qi S. Zhang, Some gradient estimates for the heat kernel equation on domains and for an equation by Perelman. Int. Math. Res. Not. (2006), Art. ID 92314, 39 pp. Department of Ma thema tics, University of Wisconsin - Madison, 480 L incoln Drive, Madison, WI 53706, U.S.A. E-mail...

2006
[1987]

ISBN: 3-540-15279-2

xii+510 pp. ISBN: 3-540-15279-2
[1988]

ISBN: 0-8176-3370-7

xii+339 pp. ISBN: 0-8176-3370-7

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.

[1] [31]

Improving Subseasonal Forecasting in the Western U.S. with Machine Learning

Shaosai Huang, Notes on Ricci ﬂows with collapsing init ial data (I): Distance distortion. Preprint, arXiv: 1809.07394

work page Pith review arXiv

[2] [10]

Je ﬀCheeger, Mikhail Gromov and Michael Taylor, Finite propaga tion speed, kernel estimates for functions of the Laplace operator, and the geometry of complete Riemannian m anifolds. J. Di ﬀerential Geom. 17 (1982), no. 1, 15-53

work page 1982

[3] [26]

Richard Hamilton, A compactness property for solution s of the Ricci ﬂow. Amer . J. Math.117 (1995), 545-572

work page 1995

[4] [41]

Preprint, arXiv: math / 0211159

Grisha Perelman, The entropy formula for the Ricci ﬂow a nd its geometric applications. Preprint, arXiv: math / 0211159

work page

[5] [16]

Mikhail Feldman, Tom Ilmanen and Lei Ni, Entropy and red uced distance for Ricci expanders. J. Geom. Anal. 15 (2005), no. 1, 49-62

work page 2005

[6] [1]

Bamler, Long-time behavior of 3-dimensional Ricci ﬂow: introduction

Richard H. Bamler, Long-time behavior of 3-dimensional Ricci ﬂow: introduction. Geom. Topol. 22 (2018) 757-774

work page 2018

[7] [2]

Besse, Einstein manifolds

Arthur L. Besse, Einstein manifolds. Ergebnisse der Mat hematik und ihrer Grenzgebiete (3), Springer-V erlag, Berlin,

work page

[8] [3]

Christoph Böhm, On the long time behavior of homogeneous Ricci ﬂows. Comment. Math. Helv. 90 (2015), no. 3, 543-571

work page 2015