Pith. sign in

REVIEW 3 major objections 2 minor 24 references

The Lipschitz truncation of functions of bounded variation

T0 review · 3 major / 2 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 This paper constructs Lipschitz approximations for functions of bounded variation that converge area-strictly while changing the function only on a small set.

desk verdict Main area-strict Lipschitz truncation for BV is real and useful, but the theorem overclaims the bad-set convergence: it is strict, not area-strict. read the letter →

arxiv 1908.10655 v1 pith:KEKS2BKC submitted 2019-08-28 math.AP

Dominic Breit , Lars Diening , Franz Gmeineder This is my paper · ORCID
classification math.AP MSC 26B3026B35
keywords functionsofboundedvariationLipschitztruncationarea-strictconvergenceLusinpropertyHardy-LittlewoodmaximalfunctionWhitneydecompositioncalculusvariationsBV
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

Functions of bounded variation (BV) are the natural setting for variational problems whose energies grow linearly in the gradient, but they cannot be approximated in the norm topology by smooth or Lipschitz functions. This paper constructs, for every $u \in BV(\Omega)$, a family of Lipschitz functions $u_\lambda$ that differ from $u$ only on a small set and converge to $u$ in the area-strict metric, the strongest convergence for which smooth approximation is still possible. Previous Lipschitz truncations, designed for Sobolev spaces, only controlled the Lebesgue measure of the changed set, and a direct transplant to BV fails: on a simple jump function the level sets of the truncation zigzag and the total variation does not converge. The new truncation corrects the bad set $\{M(Du)>\lambda\}$ by a Whitney decomposition with local mollification, restoring area-strict convergence while keeping zero boundary data. The theorem therefore gives a quantitative Lusin-type approximation that preserves the functionals and traces used in the calculus of variations.

What carries the argument

The engine is the pointwise oscillation estimate $|u(x)-u(y)|\le c|x-y|(M(Du)(x)+M(Du)(y))$ for precise representatives of BV functions, where $M$ is the Hardy-Littlewood maximal function; it makes $u$ Lipschitz on the good set $O_\lambda^\complement=\{M(Du)\le \lambda\}$. The bad set $O_\lambda$ is covered by Whitney cubes $Q_j$, and the truncation operator $T_\lambda u = u - \sum_j\big(\eta_j(u-u_j)-\varphi_j*(\eta_j(u-u_j))\big)$ replaces $u$ locally by its mean values $u_j$, then repairs each correction by convolution with a mollifier of radius $\varepsilon_j=h(\lambda)r_j/4$. The almost dual operator $S_\lambda\rho = \rho - \sum_j \eta_j(\rho-\varphi_j*\rho)$ is non-expansive on $L^\infty$ and satisfies the commutator estimate $|\langle DT_\lambda u,\rho\rangle-\langle Du,S_\lambda\rho\rangle|\le c\,h(\lambda)|Du|(O_\lambda)\|\rho\|_\infty$, which is the mechanism that turns the estimate on the bad set into area-strict convergence.

What would settle it

For $u(x)=\operatorname{sgn}(x_2-x_1)$ on $(-1,1)^2$, compute $T_\lambda u$ with an explicit Whitney covering of $O_\lambda$ and the mollifier radius $\varepsilon_j=h(\lambda)r_j/4$, then evaluate the area functional $\langle DT_\lambda u\rangle(\Omega)$. If $\limsup_{\lambda\to\infty}\langle DT_\lambda u\rangle(\Omega)>\langle Du\rangle(\Omega)=2\sqrt{2}$, the area-strict convergence claim fails; the same example shows the unmodified truncation fails by producing a zigzag with excess total variation.

Watch

Extended reading notes

Core claim

The paper's main theorem states that for $\Omega=\mathbb{R}^n$ or a bounded Lipschitz domain, given any vanishing rate $h(\lambda)\to 0$, every $u\in BV(\Omega)$ admits $u_\lambda\in W^{1,\infty}(\Omega)$ with $\|\nabla u_\lambda\|_{\infty}\le c\lambda/h(\lambda)^{n+1}$, $\{u_\lambda\neq u\}\subset O_\lambda=\{M(Du)>\lambda\}$, and $\mathcal{L}^n(O_\lambda)\le c|Du|(O_\lambda)/\lambda$. The approximations are stable in $L^q$ for $1\le q\le n/(n-1)$ and in total variation, converge area-strictly to $u$ as $\lambda\to\infty$, and preserve zero boundary values. Precisely, on the good set $O_\lambda^\complement$ the gradients $\nabla u_\lambda$ converge to the absolutely continuous part $\nabla u$ of $Du$ in $L^1$, while on the bad set they converge area-strictly to the singular part $D^s u$. In particular, the total variation and the area functional do not overshoot in the limit, which is exactly what the classical truncation cannot deliver.

Load-bearing premise

The construction rests on the pointwise oscillation estimate for BV functions, $|u(x)-u(y)|\le c|x-y|(M(Du)(x)+M(Du)(y))$ for almost every $x,y$; if this estimate were false, the good set would not be a Lipschitz set and the whole truncation would lose its control.

Editorial extensions

If this is right

  • Every BV function admits Lipschitz approximations that change the function only on a set of measure $O(|Du|(O_\lambda)/\lambda)$ and still converge area-strictly, giving a quantitative Lusin-type theorem for the BV class.
  • Because area-strict convergence implies $f$-strict convergence for every convex linear-growth integrand, integrals such as $\int_\Omega \sqrt{1+|\nabla u_\lambda|^2}\,dx$ converge to $\int_\Omega \sqrt{1+|Du|^2}$, so the truncation is compatible with minimal-surface and linear-growth energies.
  • Zero boundary data are preserved on Lipschitz domains, so Dirichlet problems for BV energies can be studied through Lipschitz approximations without changing boundary conditions.
  • The bad set carries precisely the singular part of the derivative: $\nabla u_\lambda$ on $O_\lambda^\complement$ tends to $\nabla u$ in $L^1$, while $\nabla u_\lambda\mathcal{L}^n$ on $O_\lambda$ tends area-strictly to $D^s u$.
  • Since Lipschitz functions are not norm-dense in BV, area-strict convergence is the strongest reasonable topology for this approximation problem; the theorem realizes it.

Reading between the lines

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

  • The dual-operator structure suggests a general recipe: any truncation admitting a non-expansive $L^\infty$ partner $S_\lambda$ with a commutator bound of order $o(1)|Du|(O_\lambda)$ should inherit area-strict convergence; this could be tested on parabolic or solenoidal truncations where the good-set geometry is different.
  • Choosing $h(\lambda)=\lambda^{-\alpha}$ makes the Lipschitz constant grow like $\lambda^{1+\alpha(n+1)}$ while the area excess decays like $|Du|(O_\lambda)(\lambda^{-\alpha}+\lambda^{-1})$; an explicit optimal $\alpha$ can be read off once the growth of $|Du|(O_\lambda)$ is known for a specific function, which the paper leaves implicit.
  • The failure of the classical truncation is a coarea phenomenon: its zigzag level sets lengthen the interfaces. The mollifying correction should therefore also yield convergence of the perimeters of superlevel sets $\{u>t\}$ for almost every $t$, a quantitative refinement not stated in the paper.
  • The construction uses only maximal-function oscillation estimates and Whitney geometry, so it should extend to vector-valued BV and to functions of bounded deformation, provided the analogous oscillation estimate holds; that would make the truncation available in plasticity and free-discontinuity problems.
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.

Referee Report

3 major / 2 minor

Summary. The paper constructs a Lipschitz truncation operator T_λ for functions of bounded variation, based on a Whitney covering of the bad set {M(Du)>λ} and local mollifications, and claims that T_λu approximates u area-strictly in BV(Ω), changes u only on a small set, enjoys L^q and BV stability, and admits a precise decomposition of the convergence of ∇u_λ into absolutely continuous and singular parts. The main theorem is stated for Ω=R^n and for bounded Lipschitz domains, with a separate preservation statement for zero boundary values.

Significance. If the main theorem were correct, the paper would give a genuinely useful strengthening of the classical Lipschitz truncation for BV functions: area-strict convergence instead of merely weak* or strict convergence, together with the Lusin-type small-change property. The technique of combining Whitney cubes with local mollification and the introduction of the almost-dual operator S_λ are attractive and potentially reusable ideas, and much of the error analysis (Lemmas 4, 6, 7, 9, 10) is carried out in detail. However, two independent correctness issues affect substantial parts of the main theorem, so the paper requires serious revision before its central claims can be accepted.

major comments (3)
  1. [Section 2.1, Theorem 1(d), Lemma 11(b)] The assertion that ∇u_λ L^n|_{O_λ} converges area-strictly to D^s u is not compatible with the paper's own definition of area-strict convergence. For a purely singular measure ν, equation (2.1) gives ⟨ν⟩(Ω)=L^n(Ω)+|ν|(Ω), whereas for μ_λ=∇u_λ L^n|_{O_λ} one has ⟨μ_λ⟩(Ω)=∫_{O_λ}√(1+|∇u_λ|²)dx, which converges to |D^s u|(Ω), not to L^n(Ω)+|D^s u|(Ω). The proof of Lemma 11(b) contains the same issue in the displayed chain ending with '⟨Du⟩(R^n)−⟨∇uL^n⟩(R^n)=⟨D^s u⟩(R^n)'; the left-hand side equals |D^s u|(R^n), not the area functional of D^s u. In addition, the first displayed identity in that proof has the indicator on the wrong set (it should be 1_{O_λ}∇T_λu), and the lower semicontinuity step is applied to global totals rather than to the restricted measures. These are not merely typographical: what the proof actually establishes, after the fillable weak*-lower-semicontinuity argument for the restricted measures, is strict convergence. The statement of Theorem 1(d) should therefore read that ∇u_λL^n|_{O_λ} converges strictly to D^s u, and Lemma 11(b) should be relabelled accordingly.
  2. [Section 3.1, Eq. (3.1), Theorem 1(c)-(e)] The construction does not handle the boundary of a bounded Lipschitz domain for general BV data. The proof extends u to R^n by zero, so the derivative of the extended function has boundary concentration whenever the trace of u is nonzero. For Whitney cubes with 3/4 Q_j not contained in Ω, definition (3.1) sets u_j=0, which forces the truncation to interpolate toward zero near ∂Ω. For the example u≡1 on a bounded Lipschitz domain, this produces a boundary layer of width proportional to h(λ)/λ and gradient of size λ/h(λ), so the L^1 norm of ∇u_λ inside Ω contains a positive contribution that does not vanish as λ→∞, while |Du|(Ω)=0. Thus the claimed bound in Theorem 1(c), ‖∇u_λ‖_{L^1(Ω)}≤c|Du|(Ω), and the claimed area-strict convergence in Theorem 1(d) fail for this construction. Lemma 6 only bounds the total variation on all of R^n in terms of |D(u~)|(R^n), which includes the boundary trace, and the proof in Section 3.5 does not close this gap. The boundary treatment in Section 3.4 is appropriate for the zero-boundary case (e), but the general bounded-domain case is not established.
  3. [Section 2.1, Theorem 1 for Ω=R^n] Area-strict convergence is defined via ⟨μ_k⟩(Ω)→⟨μ⟩(Ω), but for Ω=R^n the quantity ⟨μ⟩(R^n) is infinite for any finite nontrivial measure if the definition in (2.1) is applied literally, because the term f(0)L^n=R^n contributes an infinite Lebesgue-volume part. Consequently, in the Ω=R^n case the convergence condition in the definition is vacuous in its Lebesgue part, and inequalities such as (3.17) in Lemma 10 are formal. The manuscript should either restrict the area-strict statements to bounded domains or give a precise local (or compactly supported) definition of area-strict convergence for unbounded domains and adapt the proof of Lemma 10 accordingly.
minor comments (2)
  1. [Lemma 6, Eq. (3.5)] The first inequality in (3.5), ‖B_ju‖_q ≤ c∫_{Q_j}|u|dx, is not correct as stated: it is not homogeneous when q>1 and fails for functions concentrated on a small subset of Q_j. The proof immediately establishes the q-th power form ∫|B_ju|^q dx ≤ c∫_{Q_j}|u|^q dx, which is the version used in (3.6). The display should be corrected.
  2. [Lemma 11(b), first displayed formula] In the proof of Lemma 11(b), the first identity should pair the test function with 1_{O_λ}∇T_λu, not with 1_{O_λ^c}∇T_λu; as written the formula proves convergence of the good-set restriction, which is the content of part (a), and does not address the bad-set restriction.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the Lipschitz truncation and its area-strict convergence are derived from independent estimates, not assumed as inputs.

full rationale

The paper constructs uλ = Tλu by a Whitney-cover correction and proves the main theorem from self-contained ingredients: Lemma 2(c) is a pointwise oscillation estimate quoted from DeVore–Sharpley and proved in the text; Lemma 6 and Lemma 7 give the stability and Lipschitz bounds; Lemma 9 gives the commutator estimate (3.14) with a direct proof. The function h(λ) is an arbitrary quantifier in the statement and enters the mollifier radius; it is not fitted to data, and the bad set Oλ is defined from the original function's maximal function, not from the approximants. The conclusions of Theorem 1(d) are not used in the construction: Lemma 10 first proves global area-strict convergence, and Lemma 11 then identifies the good-set and bad-set limits. The self-citations (e.g., DKS13, DSSV17, DRW10) provide auxiliary Whitney-covering or stability lemmas that are either reproved in the text or are standard; no uniqueness theorem is imported to force the ansatz. A separate correctness concern is present in Lemma 11(b): the proof gives a limsup bound for ⟨DTλu⟩(Oλ) and then invokes global lower semicontinuity, which does not by itself yield the required liminf on the restricted bad set; this is an unproven step or possible gap in the argument, not a circular reduction. Since no prediction or derived claim reduces to its own input by construction, the circularity score is 0.

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

No free parameters or invented entities. The construction depends on standard covering, maximal-function, and lower-semicontinuity tools. The function h(λ) is a quantifier in the theorem statement, not a fitted constant, and the theorem holds for any admissible h.

assumptions (5)
  • standard math BV extension by zero and Poincaré inequalities for BV functions
    Used in Lemma 4 and in the reduction from bounded Lipschitz domains to R^n at the start of Section 3.
  • standard math Whitney decomposition with geometric alternatives (Lemma 3)
    A known covering lemma for the bad set O_λ, used throughout the proof, especially in Lemma 7.
  • standard math DeVore-Sharpley maximal function estimate and the pointwise oscillation bound for BV
    Underpins Lemma 2(c), the key estimate showing BV functions are Lipschitz on the good set.
  • domain assumption Outer cone condition for Lipschitz domains, equation (3.21)
    Needed in Section 3.4 to define directed mollifiers that preserve zero boundary values.
  • standard math Lower semicontinuity of convex area-type functionals under weak* convergence (Goffman-Serrin, Reshetnyak)
    Invoked in (2.3) and used in Lemmas 10 and 11 to identify limits in the area-strict topology.

how reviews work

0 comments
Cite this review

Pith. "Pith review of The Lipschitz truncation of functions of bounded variation." pith.science (2026). https://pith.science/paper/KEKS2BKC

@misc{pith2026190810655,
  author       = {Pith},
  title        = {Pith review of: The Lipschitz truncation of functions of bounded variation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/KEKS2BKC}},
  note         = {Machine review of arXiv:1908.10655}
}
read the original abstract

We construct a Lipschitz truncation which approximates functions of bounded variation in the area-strict metric. The Lipschitz truncation changes the original function only on a small set similar to Lusin's theorem. Previous results could only give estimates on the Lebesgue measure of the set where the Lipschitz approximations differ from the original function.

Figures

Figures reproduced from arXiv: 1908.10655 by the authors.

Figure 1
Figure 1. Example of Remark 5. Left: The picture shows a zoom of the Whitney cover at the diagonal. The numbers indicate the mean values (u)Qj . The function uλ is locally constant outside the shaded region. Right: The picture shows the resulting isolines of uλ. The special choice of this truncation operator will become clear later when we consider its (almost) dual operator in Lemma 9. As we will see, the map Tλu defines an … view at source ↗
Figure 2
Figure 2. Local chart of the boundary. The cones indicate the direction of the convolution. has to replace ϕj in the definition of the (almost) dual operator Sλ by ϕj with ϕj (x) := ϕj (−x). 3.5. Proof of Theorem 1. We are now in position to prove our main theorem. For a given λ > 0, define uλ := Tλu as in (3.4). The Lipschitz property (a) follows from Corollary 8. The smallness of the set {u 6= uλ} from (b) is an immediate c… view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

24 extracted references · 24 canonical work pages

  1. [1]

    Rational Mech

    Emilio Acerbi and Nicola Fusco, Semicontinuity problems in the calculus of variations, Arch. Rational Mech. Anal. 86 (1984), no. 2, 125--145

    work page 1984

  2. [2]

    Rational Mech

    , A regularity theorem for minimizers of quasiconvex integrals, Arch. Rational Mech. Anal. 99 (1987), no. 3, 261--281

    work page 1987

  3. [3]

    Acerbi and N

    E. Acerbi and N. Fusco, An approximation lemma for W 1,p functions , Material instabilities in continuum mechanics (Edinburgh, 1985--1986), Oxford Sci. Publ., Oxford Univ. Press, New York, 1988, pp. 1--5

    work page 1985

  4. [4]

    Ambrosio, N

    L. Ambrosio, N. Fusco, and D. Pallara, Functions of bounded variation and free discontinuity problems, Oxford Mathematical Monographs, The Clarendon Press Oxford University Press, New York, 2000

    work page 2000

  5. [5]

    Breit, L

    D. Breit, L. Diening, and M. Fuchs, Solenoidal L ipschitz truncation and applications in fluid mechanics , J. Differential Equations 253 (2012), no. 6, 1910--1942

    work page 2012

  6. [6]

    PDE 9 (2016), no

    Miroslav Bul\' c ek, Lars Diening, and Sebastian Schwarzacher, Existence, uniqueness and optimal regularity results for very weak solutions to nonlinear elliptic systems, Anal. PDE 9 (2016), no. 5, 1115--1151

    work page 2016

  7. [7]

    Menita Carozza, Nicola Fusco, and Giuseppe Mingione, Partial regularity of minimizers of quasiconvex integrals with subquadratic growth, Ann. Mat. Pura Appl. (4) 175 (1998), 141--164

    work page 1998

  8. [8]

    Lars Diening, Christian Kreuzer, and Endre S\" u li, Finite element approximation of steady flows of incompressible fluids with implicit power-law-like rheology, SIAM J. Numer. Anal. 51 (2013), no. 2, 984--1015

    work page 2013

Show all 24 references
  1. [9]

    Diening, D

    L. Diening, D. Lengeler, B. Stroffolini, and A. Verde, Partial R egularity for M inimizers of Q uasi-convex F unctionals with G eneral G rowth , SIAM J. Math. Anal. 44 (2012), no. 5, 3594--3616

    2012

  2. [10]

    Diening, J

    L. Diening, J. M \'a lek, and M. Steinhauer, On L ipschitz truncations of S obolev functions (with variable exponent) and their selected applications , ESAIM Control Optim. Calc. Var. 14 (2008), no. 2, 211--232

    2008

  3. [11]

    Lars Diening, Michael R u z i c ka, and J \"o rg Wolf, Existence of weak solutions for unsteady motions of generalized N ewtonian fluids , Ann. Sc. Norm. Super. Pisa Cl. Sci. (5) 9 (2010), no. 1, 1--46

    2010

  4. [12]

    DeVore and Robert C

    Ronald A. DeVore and Robert C. Sharpley, Maximal functions measuring smoothness, Mem. Amer. Math. Soc. 47 (1984), no. 293, viii+115

    1984

  5. [13]

    Diening, S

    L. Diening, S. Schwarzacher, B. Stroffolini, and A. Verde, Parabolic L ipschitz truncation and caloric approximation , Calc. Var. Partial Differential Equations 56 (2017), no. 4, Art. 120, 27

    2017

  6. [14]

    Evans and Ronald F

    Lawrence C. Evans and Ronald F. Gariepy, Measure theory and fine properties of functions, Studies in Advanced Mathematics, CRC Press, Boca Raton, FL, 1992. 1158660

    1992

  7. [15]

    Frehse, J

    J. Frehse, J. M \'a lek, and M. Steinhauer, On analysis of steady flows of fluids with shear-dependent viscosity based on the L ipschitz truncation method , SIAM J. Math. Anal. 34 (2003), no. 5, 1064--1083 (electronic)

    2003

  8. [16]

    Casper Goffman and James Serrin, Sublinear functions of measures and variational integrals, Duke Math. J. 31 (1964), 159--178

    1964

  9. [17]

    R\" u diger Landes, Quasimonotone versus pseudomonotone, Proc. Roy. Soc. Edinburgh Sect. A 126 (1996), no. 4, 705--717

    1996

  10. [18]

    Lewis, On very weak solutions of certain elliptic systems, Comm

    John L. Lewis, On very weak solutions of certain elliptic systems, Comm. Partial Differential Equations 18 (1993), no. 9-10, 1515--1537

    1993

  11. [19]

    Fon Che Liu, A L uzin type property of S obolev functions , Indiana Univ. Math. J. 26 (1977), no. 4, 645--651

    1977

  12. [20]

    Lusin , Sur les propri\'et\'es des fonctions mesurables

    N. Lusin , Sur les propri\'et\'es des fonctions mesurables. , C. R. Acad. Sci., Paris 154 (1912), 1688--1690 (French)

    1912

  13. [21]

    Mal \'y and W

    J. Mal \'y and W. P. Ziemer, Fine regularity of solutions of elliptic partial differential equations, American Mathematical Society, Providence, RI, 1997

    1997

  14. [22]

    Ju. G. Re s etnjak, The weak convergence of completely additive vector-valued set functions, Sibirsk. Mat. Z . 9 (1968), 1386--1394

    1968

  15. [23]

    Süli and T

    E. Süli and T. Tscherpel, Fully discrete finite element approximation of unsteady flows of implicitly constituted incompressible fluids, IMA Journal of Numerical Analysis (2019)

    2019

  16. [24]

    1306, Springer, Berlin, 1988, pp

    Ke-Wei Zhang, On the D irichlet problem for a class of quasilinear elliptic systems of partial differential equations in divergence form , Partial differential equations ( T ianjin, 1986), Lecture Notes in Math., vol. 1306, Springer, Berlin, 1988, pp. 262--277

    1986

Pith tools

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