Existence of curvature flow with forcing in a critical Sobolev space

Yoshihiro Tonegawa; Yuning Liu

REVIEW 3 major objections 4 minor 19 references

Existence of curvature flow with forcing in a critical Sobolev space

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 The paper proves that a curve in the plane can be moved by the sum of its curvature and an external vector field even when the forcing field has only critical Sobolev regularity, and that the resulting weak flow exists globally and passes…

desk verdict Critical-case existence for 1D curvature flow with forcing is plausibly proved; the a priori estimates are solid, but the appendix proving the smooth-forcing input is a sketch and needs to be filled before the result is fully established. read the letter →

arxiv 2411.18284 v2 pith:HFXHVVR3 submitted 2024-11-27 math.AP math.DG

Yuning Liu , Yoshihiro Tonegawa This is my paper · ORCID

classification math.APmath.DG MSC 53E1049Q15

keywords meancurvatureflowBrakkecriticalSobolevspacevarifoldcurveshorteningforcingtermmulti-phasesetsoffiniteperimeter

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 shows that a one-dimensional curve in the plane has a global-in-time weak solution to the motion law "normal velocity equals curvature plus an external vector field," even when the vector field is only in the dimensionally critical Sobolev class $L^\infty_{\mathrm{loc}}([0,\infty);L^2(\mathbb{R}^2)) \cap L^2_{\mathrm{loc}}([0,\infty);W^{1,2}(\mathbb{R}^2))$. This is the borderline regularity at which the forcing term has the same strength as curvature under parabolic scaling, so the problem is genuinely critical rather than a perturbation of ordinary mean curvature flow. The solution is a Brakke-type varifold flow together with evolving sets of finite perimeter, it starts from any closed rectifiable initial set of finite length, and it is non-trivial for at least a positive time interval. The paper also gives an explicit bound on the growth of length in terms of the initial length and a scale-invariant norm of the forcing field.

What carries the argument

The argument is carried by a priori estimates for one-dimensional varifolds with $L^2$ curvature, combined with a measure-gradient inequality (Theorem 3.2) and a Gronwall argument in Proposition 3.3. The density bound $D(\|V\|) \le (\|V\|(\mathbb{R}^2))^{1/2}(\int |h(V)|^2\,d\|V\|)^{1/2}$ lets the paper control the integral of $|u|^2$ against the varifold measure by the $L^2$ norm of curvature and the $W^{1,2}$ norm of $u$. These estimates replace the monotonicity formula that handles subcritical forcings and yield the explicit bounds (3.6)-(3.8). The approximating flows are supplied by a smooth-forcing multi-phase Brakke flow theorem (Theorem 4.2), obtained by modifying known constructions, and the final solution is extracted as a limit after mollifying the vector field $u$.

What would settle it

Compute, for a fixed smooth compactly supported vector field $u$ and a circular initial curve, the length at time $T$ of the classical solution to (1.1) and compare it with the bound $\mathcal{H}^1(\Gamma(T)) \le \mathcal{H}^1(\Gamma_0)\exp(c_2^2 c_1(u,T))$; any smooth datum violating this inequality would disprove the a priori estimate behind the theorem. Alternatively, exhibit a closed rectifiable $\Gamma_0$ and a $u$ in class (1.2) for which the mollified flows produced by Theorem 4.2 have no convergent subsequence of measures on a dense time set, contradicting Proposition 5.1.

Watch

Extended reading notes

Core claim

Under Assumptions (A1)-(A4), there exist a family of varifolds $\{V_t\}$ and sets of finite perimeter $\{E_i(t)\}$ such that the Brakke inequality (2.11) holds with normal velocity $h + u^\perp$, the length bound (2.10) holds, and the flow has the structural properties (V3) and (E5)-(E8): at almost every time it is a finite union of $W^{2,2}$ embedded curves meeting at junctions with angles $0$, $60$, or $120$ degrees, and the phase boundaries are controlled in a measure-theoretic sense. This is a genuine critical-case existence theorem: it covers every closed $1$-rectifiable $\Gamma_0$ with $\mathcal{H}^1(\Gamma_0)<\infty$ and every $u$ in the critical class (1.2), and the solution exists for all time while allowing singularities.

Load-bearing premise

The proof depends on the assertion in Theorem 4.2 that the known multi-phase Brakke flow construction still works when a smooth forcing term is added to the velocity; Appendix A only sketches the necessary modifications, with several steps left as "identical" or "one can deduce". If any of those unsupplied modifications fails, the mollified flows used in the limiting argument do not exist and the main theorem is not established.

Editorial extensions

If this is right

For two-phase initial data, the flow consists almost everywhere of embedded closed curves that may only intersect tangentially, so no triple junctions appear in the $N=2$ case.
The explicit length bound $\mathcal{H}^1(\Gamma(T)) \le \mathcal{H}^1(\Gamma_0)\exp(c_2^2 c_1(u,T))$ holds for all $T>0$, so the curve length remains finite at every time for any forcing field with finite $c_1(u,T)$.
The phase sets $E_i(t)$ are $1/2$-H\"older continuous in time with respect to Lebesgue measure and satisfy the estimate (E5), so they cannot suddenly vanish.
When the flow has unit density for almost every time, it coincides with the reduced boundaries of the phase sets: $V_t = \mathrm{var}(\bigcup_i \partial^*E_i(t),1)$.
Because the forcing and curvature have comparable strength at this critical regularity, no regularity theorem of the kind available in the subcritical case is known for the resulting flow.

Reading between the lines

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

The same a priori estimates should extend to higher-dimensional surfaces, since the paper states its smooth-forcing existence theorem in general dimension but only proves the critical existence result for $n=1$.
The critical Sobolev class (1.2) is the natural regularity class for 2D Navier-Stokes flows, so the estimates here are a step toward a coupled two-phase Navier-Stokes/mean-curvature problem in which $u$ is no longer a fixed datum; the paper leaves that coupling for future work.
The absence of a regularity theorem in the critical case suggests that singularities may occur on a dense set of times; a natural test is whether the constructed flow satisfies any partial regularity property analogous to the subcritical case.

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

Here's my take on Liu–Tonegawa. The critical case p=q=n+1=2 was explicitly open, since the subcritical condition (1.5) fails exactly there. The paper's strategy is sound: replace Huisken's monotonicity with the 1D varifold density estimate (Prop 3.1), get Gronwall control on length, L2 curvature, and the trace of u, then mollify u and pass to the limit. Section 3 is the heart and it is rigorous. The limit argument in Section 5 is mostly standard and the error terms in Prop 5.5 are handled carefully. If the smooth-forcing input holds, the main theorem follows.

The soft spot is that input. Theorem 4.2 is proved by reference and sketch in Appendix A. The authors state the key new estimates (A.1)–(A.4), but they do not define the modified class of volume-controlled deformations Evc, the verification of (A.7)–(A.9) is not written out, and the modified Huisken-type monotonicity (A.22) is asserted after a short computation. Lemma A.1 is proved, but it rests on that computation. These are load-bearing: without Theorem 4.2 there are no approximate flows. I don't think the gaps are fatal—they look fillable by someone who knows the Kim–Tonegawa and Stuvard–Tonegawa machinery—but they are not cosmetic. The paper would be much stronger if the appendix were expanded into a complete proof, or if Theorem 4.2 were separated out with a full proof.

The citation pattern is fine. [8,9,17] are the prior constructions, and the critical-case argument does not depend on the main theorem. No circularity.

Bottom line: this is a significant within-field result and deserves a serious referee. I would send it out, expecting major revision to close the appendix gaps.

Referee Report

3 major / 4 minor

Summary. The paper proves a global-in-time existence theorem for the motion of a one-dimensional interface in R^2 with normal velocity h + u^⊥, where h is curvature and u is a forcing vector field in the critical Sobolev class (1.2). The main result, Theorem 2.2, asserts that under assumptions (A1)-(A4) there exist a varifold flow {V_t} and a family of finite-perimeter sets {E_i(t)} satisfying the Brakke-type inequality (2.11), the length bound (2.10), structural regularity (V3), and the phase properties (E5)-(E8). The proof strategy is to mollify u, apply a smooth-forcing existence result (Theorem 4.2) obtained by modifying the multi-phase Brakke flow constructions of Kim-Tonegawa and Stuvard-Tonegawa, derive uniform a priori estimates in Section 3, and pass to the limit in Section 5. The a priori estimates and the limiting argument are largely explicit, while the proof of Theorem 4.2 is delegated to Appendix A, where many steps are only sketched.

Significance. The result is significant because it extends existing subcritical existence results for mean curvature flow with transport term to the dimensionally critical case p = q = n+1 = 2, where the forcing term is no longer a small perturbation. The quantitative a priori estimates in Section 3, especially the length bound (3.6) and the curvature and trace estimates (3.7)-(3.8), are clean and self-contained, and the limiting argument in Section 5 is mostly detailed. However, the paper's central claim depends on Theorem 4.2 for smooth forcing, whose proof is only sketched in Appendix A; the volume-control condition and the forced monotonicity formula are asserted rather than verified. Since these are load-bearing for the existence of the approximate flows, the paper is not yet fully convincing as written.

major comments (3)

[Appendix A, around (A.7)-(A.9) and (A.11)] The proof of Theorem 4.2 is the load-bearing base case for the whole paper, but the modifications to [8,9,17] are not fully demonstrated. In particular, the volume-controlled Lipschitz deformation class E_vc from [17, Definition 3.1] is referenced but not restated for the forced problem, and the map f_2(x) = x + Δt(h_ε(x,∂E*) + u(x,(l+1)Δt)) in (A.11) is not shown to satisfy the volume-control estimates needed in [17]. The transition from (A.1)-(A.3) to (A.7)-(A.9) is justified by 'the rest of the proof is identical and one can deduce', which is not sufficient for a theorem on which the main result collapses if any modification fails. Please provide the explicit verification of the E_vc condition with the extra u-term, or state precisely which lemmas of [8,17] are invoked and how the new terms are controlled.
[Appendix A, (A.21)-(A.22)] The modified Huisken monotonicity inequality (A.22) is asserted after replacing the Gaussian weight by exp(-t||u||_{L^∞}^2) ρ, with the statement that 'proceeding as in [7, Proposition 6.2]' yields the estimate. This is delicate because u appears in the motion law and generates additional terms involving u and its derivatives in the evolution of the weighted measure; the exponential factor must compensate for these terms. Since (A.22) is used, via [8, Sections 7-8], to control density ratios and to prove rectifiability and integrality, a failure here would invalidate (V3') and the structural properties of the limiting flow. A full derivation of (A.22), or a precise reference to a published proof of the forced monotonicity inequality, is needed.
[Proposition 5.3, case (b)] The proof that non-line tangent cones are discrete and that the limiting junction angles are exactly 0, 60, or 120 degrees is sketched: the text says 'one can argue that there are a finite number of W^{2,2} curves reaching to the junction point' and 'the angles of intersection have to be either 0, 60 or 120 degrees due to the fact that V_s^{(m'_j)} is converging with the same property of junctions.' Since property (V3) is part of the main theorem and this is the only proof of the junction structure, the argument should be expanded, in particular the compactness of the set of junction points and the angle classification in the limit.

minor comments (4)

[Proposition 5.10] The text says 'the density of ||V_t|| is precisely the number of W^{1,1} curves passing through that point'; this should presumably be 'W^{2,2} curves' to match the regularity established in (V3).
[Theorem 4.2, (V3')] The statement 'if n = 1, (V3) of Theorem 2.2 holds' is a forward reference that forces the reader to compare theorems; it would be clearer to state the one-dimensional structural property directly in Theorem 4.2.
[Proposition 5.5, equation (5.23b)] In the displayed formula after (5.23a), the term 'h(V_s)·(∇φ − gφ)' mixes notation: the first product is a vector dot product while the preceding '∇φ · g^⊥' is also a vector dot product, but the expression would benefit from consistent placement of parentheses to avoid ambiguity.
[Footnote in Proposition 5.2] The footnote explains that at exceptional times V_s is defined using a fixed S_0 and is not rectifiable; using the same symbol V_s for these non-rectifiable varifolds is slightly confusing, though the measure of such times is zero.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the critical forcing theorem is obtained by a standard mollification-and-limit argument from prior independent smooth-forcing existence results, with no fitted parameters or definitional equivalences.

full rationale

The derivation chain is not circular. Theorem 2.2 is proved by mollifying the critical-Sobolev forcing u to smooth u^(m), applying Theorem 4.2 for smooth-forcing Brakke flows, deriving m-independent estimates via Proposition 3.3, and passing to a limit via Allard compactness and BV compactness. None of the claimed target properties (V1)-(V8), (E1)-(E8) appear as assumptions in the inputs: the smooth-forcing theorem concerns C^1_c u, while the target u is only in L^2_t W^{1,2}_x, and the limit passage supplies the critical case. The a priori estimates are genuine energy estimates, not fitted constants. The main self-citation load is Theorem 4.2, which is adapted from [8,9,17] (both involving the second author). Those prior works are independent published results that do not contain the critical target theorem, so citing them is real evidence rather than a circular reduction. One completeness caveat exists but is not circularity: Appendix A states 'The rest of the proof is identical and one can deduce (A.7)-(A.9)' and 'we can proceed exactly as in [8, Section 10]', so several modifications (e.g., membership of x+(h_eps+u)Delta t in the volume-controlled deformation class, the forced monotonicity estimate (A.22)) are left as sketches. If those verifications fail, Theorem 4.2 is unsupported and the limit argument has no starting point; that is a proof-completeness risk, not an equivalence between theorem and input. Score 0 reflects the absence of circular reduction.

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

The central claim introduces no free parameters or invented entities. It relies on standard geometric measure theory, the Meyer-Ziemer inequality, and, most importantly, on a prior existence theorem for smooth forcing (Theorem 4.2) whose proof is only sketched via modifications to [8,9,17].

assumptions (5)

standard math Standard geometric measure theory background (varifolds, first variation, Allard compactness, Brakke perpendicularity)
Used throughout Sections 2-5; cited to [2,3,15].
standard math Meyer-Ziemer inequality (Theorem 3.2), giving ∫ φ dµ ≤ c2 D(µ) ∫ |∇φ|
Quoted with proof reference [13]; essential for the a priori estimates in Proposition 3.3.
domain assumption Existence of multi-phase Brakke flows for smooth forcing (Theorem 4.2), obtained by modifying Kim-Tonegawa and Stuvard-Tonegawa [8,9,17]
The proof is only sketched in Appendix A; the main theorem depends on it.
domain assumption Initial data assumptions (A1)-(A4): partition of R^2 by N phases with rectifiable finite-length boundary, u in critical Sobolev class
These are the hypotheses of the theorem, not derived.
standard math One-dimensional varifold monotonicity-type estimates (Proposition 3.1), following Pozzetta [14]
Used to bound the density ratio D(||V||) in terms of ||δV|| and L2 curvature.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Existence of curvature flow with forcing in a critical Sobolev space." pith.science (2026). https://pith.science/paper/HFXHVVR3

@misc{pith2026241118284,
  author       = {Pith},
  title        = {Pith review of: Existence of curvature flow with forcing in a critical Sobolev space},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/HFXHVVR3}},
  note         = {Machine review of arXiv:2411.18284}
}

abstract

Suppose that a closed 1-rectifiable set $\Gamma_0\subset\mathbb R^2$ of finite 1-dimensional Hausdorff measure and a vector field $u$ in a dimensionally critical Sobolev space are given. It is proved that, starting from $\Gamma_0$, there exists a non-trivial flow of curves with the normal velocity given by the sum of the curvature and the given vector field $u$. The motion law is satisfied in the sense of Brakke and the flow exists through singularities.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

19 extracted references · 18 canonical work pages

[17]

Stuvard and Y

S. Stuvard and Y. Tonegawa. On the existence of canonica l multi-phase Brakke ﬂows. Adv. Calc. Var., 17(1):33–78, 2024

work page 2024
[1]

R. A. Adams and J. J. F. Fournier. Sobolev spaces, volume 140 of Pure and Applied Mathematics (Amsterdam). Elsevier/Academic Press, Amsterdam, second edition, 200 3
[2]

W. K. Allard. On the ﬁrst variation of a varifold. Ann. Math., 95(3):417–491, 1972

work page 1972
[3]

K. A. Brakke. The Motion of a Surface by Its Mean Curvature. , volume 20 of Mathematical notes . Princeton University Press, Princeton, 1978

work page 1978
[4]

Caﬀarelli, R

L. Caﬀarelli, R. Kohn, and L. Nirenberg. Partial regular ity of suitable weak solutions of the Navier- Stokes equations. Comm. Pure Appl. Math. , 35(6):771–831, 1982

work page 1982
[5]

L. C. Evans and R. F. Gariepy. Measure theory and ﬁne properties of functions . Textbooks in Mathe- matics. CRC Press, revised edition, 2015. 24 Y. LIU AND Y. TONEGA W A

work page 2015
[6]

G. Huisken. Asymptotic behavior for singularities of th e mean curvature ﬂow. J. Diﬀerential Geom. , 31(1):285–299, 1990

work page 1990
[7]

Kasai and Y

K. Kasai and Y. Tonegawa. A general regularity theory for weak mean curvature ﬂow. Calc. Var. Partial Diﬀer. Equ. , 50(1):1–68, 2014

work page 2014

Show all 19 references

[8]

Kim and Y

L. Kim and Y. Tonegawa. On the mean curvature ﬂow of grain b oundaries. Ann. Inst. Fourier (Greno- ble), 67(1):43–142, 2017

2017
[9]

Kim and Y

L. Kim and Y. Tonegawa. Existence and regularity theorem s of one-dimensional Brakke ﬂows. Inter- faces Free Bound., 22(4):505–550, 2020

2020
[10]

C. Liu, N. Sato, and Y. Tonegawa. On the existence of mean curvature ﬂow with transport term. Interfaces Free Bound., 12(2):251–277, 2010

2010
[11]

C. Liu, N. Sato, and Y. Tonegawa. Two-phase ﬂow problem c oupled with mean curvature ﬂow. Inter- faces Free Bound., 14(2):185–203, 2012

2012
[12]

F. Maggi. Sets of ﬁnite perimeter and geometric variational problems : an introduction to Geometric Measure Theory, volume 135 of Cambridge Studies in Advanced Mathematics . Cambridge University Press, Cambridge, 2012

2012
[13]

N. G. Meyers and W. P. Ziemer. Integral inequalities of P oincaré and Wirtinger type for BV functions. Amer. J. Math. , 99(6):1345–1360, 1977

1977
[14]

Pozzetta

M. Pozzetta. A varifold perspective on the p-elastic energy of planar sets. J. Convex Anal. , 27(3):845– 879, 2020

2020
[15]

L. Simon. Lectures on geometric measure theory, volume 3 of Proceedings of the Centre for Mathemat- ical Analysis. Australian National University, Canberra, 1983

1983
[16]

Stuvard and Y

S. Stuvard and Y. Tonegawa. An existence theorem for Bra kke ﬂow with ﬁxed boundary conditions. Calc. Var. Partial Diﬀer. Equ. , 60(1):Paper No. 43, 53, 2021

2021
[18]

Takasao and Y

K. Takasao and Y. Tonegawa. Existence and regularity of mean curvature ﬂow with transport term in higher dimensions. Math. Ann. , 364(3):857–935, 2016

2016
[19]

Tonegawa

Y. Tonegawa. Brakke’s Mean Curvature Flow: An Introduction . SpringerBriefs in Mathematics. Springer, Singapore, 2019. NYU Shanghai, 567 Yangsi W road, Pudong, Shanghai 200126, Ch ina Email address : yl67@nyu.edu Department of Mathematics, Institute of Science Tokyo, 2-1 2-1 O...

2019

Pith tools

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

[1] [17]

Stuvard and Y

S. Stuvard and Y. Tonegawa. On the existence of canonica l multi-phase Brakke ﬂows. Adv. Calc. Var., 17(1):33–78, 2024

work page 2024

[2] [1]

R. A. Adams and J. J. F. Fournier. Sobolev spaces, volume 140 of Pure and Applied Mathematics (Amsterdam). Elsevier/Academic Press, Amsterdam, second edition, 200 3

[3] [2]

W. K. Allard. On the ﬁrst variation of a varifold. Ann. Math., 95(3):417–491, 1972

work page 1972

[4] [3]

K. A. Brakke. The Motion of a Surface by Its Mean Curvature. , volume 20 of Mathematical notes . Princeton University Press, Princeton, 1978

work page 1978

[5] [4]

Caﬀarelli, R

L. Caﬀarelli, R. Kohn, and L. Nirenberg. Partial regular ity of suitable weak solutions of the Navier- Stokes equations. Comm. Pure Appl. Math. , 35(6):771–831, 1982

work page 1982

[6] [5]

L. C. Evans and R. F. Gariepy. Measure theory and ﬁne properties of functions . Textbooks in Mathe- matics. CRC Press, revised edition, 2015. 24 Y. LIU AND Y. TONEGA W A

work page 2015

[7] [6]

G. Huisken. Asymptotic behavior for singularities of th e mean curvature ﬂow. J. Diﬀerential Geom. , 31(1):285–299, 1990

work page 1990

[8] [7]

Kasai and Y

K. Kasai and Y. Tonegawa. A general regularity theory for weak mean curvature ﬂow. Calc. Var. Partial Diﬀer. Equ. , 50(1):1–68, 2014

work page 2014