The Endpoint Cardinality of Discrete Cube Skeleta

Dean Menezes

The Endpoint Cardinality of Discrete Cube Skeleta

T0 review · 0 major / 6 minor · reviewed 2026-08-01 · deepseek-v4-flash

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

Pith's one-line read This paper determines the exact minimum size of a lattice set that carries N filled cube k-skeleta.

desk verdict The paper closes the open endpoint lower bound for lattice cube skeleta with a genuinely new labelled-entropy/balanced-induction argument, and the proof is sound. read the letter →

arxiv 2607.15502 v1 pith:ONGI6FMD submitted 2026-07-16 math.CO cs.ITmath.ITmath.MG

Dean Menezes This is my paper · ORCID

classification math.COcs.ITmath.ITmath.MG MSC 05D0552C10

keywords cubeskeletondiscretegeometryentropyShearer'sinequalitylatticesetsquareboundariesendpointexponentfinitecardinality

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 establishes that the minimum size of a finite lattice set containing a filled axis-parallel discrete k-skeleton about every point of an N-point center set is, up to constants depending only on n and k, exactly N^{1-(n-k)/(2n^2)}. Earlier work produced constructions at this exponent and lower bounds at every smaller exponent, but the matching lower bound at the endpoint was open for k≥1; this paper closes that gap. The main interest is that the exponent is exact and that the proof avoids the logarithmic losses of previous approaches. In the planar case n=2, k=1 (square boundaries), the result says the carrier must have at least cN^{7/8} points, and this exponent is sharp.

What carries the argument

The central objects are the filled discrete k-skeleton S_k(x,r) (all points whose coordinates coincide with x within r on some k coordinates and equal x±r on the rest) and the cofactor estimate, which uses a labelled form of Shearer's projection inequality to convert cube-vertex entropy into a lower bound on the number of distinct (n−k)-tuples of fixed coordinates. The proof's load-bearing mechanism is a strong induction that balances two regimes: a large-radius branch where faces of length at least aN^{1/n} are disjoint, and a small-radius branch where the integer lattice confines at most h^n centers to each cell, giving a contraction that closes the induction.

What would settle it

For n=2,k=1, try to build an infinite sequence of finite sets B_N⊂Z^2 with |B_N| ≤ C N^{7/8−ε} for some ε>0 that still contain a square boundary about each of N distinct lattice centers. If such a sequence existed, Corollary 1.3 would be false; the paper's construction, by contrast, matches the O(N^{7/8}) upper bound.

Watch

Extended reading notes

Core claim

For fixed 0≤k<n, define F_{n,k}(N) as the least size of a set B⊂Z^n that contains a filled discrete k-skeleton S_k(x,r_x) about each of N centers x with integral radii. The paper proves two-sided bounds c N^{1-(n-k)/(2n^2)} ≤ F_{n,k}(N) ≤ C N^{1-(n-k)/(2n^2)}. The lower bound is the main work: it combines a cube-vertex midpoint estimate with a labelled uniform-cover entropy inequality to show that many labeled normal cofactors must be distinct, then a strong induction splits centers into those with radius at least aN^{1/n} and those with smaller radius. Large radii give many disjoint faces; small radii are confined to lattice cells with fewer than N centers, so induction closes with the same

Load-bearing premise

The argument requires the centers to live on the integer lattice Z^n, so that every cell of side h contains at most h^n centers and the small-radius branch can invoke induction on fewer centers; without this lattice occupancy bound, the endpoint lower bound is not established.

Editorial extensions

If this is right

Corollary 1.3: for n=2,k=1, any lattice set containing a square boundary about each of N centers has at least cN^{7/8} points, and this exponent is attained.
For every fixed n,k, the exponent β=1−(n−k)/(2n^2) is the true growth rate, so no further improvement in the power of N is possible.
The lower bound holds for all N uniformly with constants depending only on n and k, not on the configuration.
The method eliminates the logarithmic factor that appeared in earlier near-endpoint lower bounds for the planar boundary case.
The construction in Section 5 provides an efficient carrier of size O(N^β), so the theorem is two-sided.

Reading between the lines

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

The proof avoids any dyadic pigeonhole step; the same balanced local-global induction may transfer to other combinatorial incidence problems where large and small scales compete, yielding log-free bounds.
The lattice assumption is essential: for centers in R^n, local cell occupancy fails; a natural testable extension is to replace Z^n by a separated set in R^n and check whether the same exponent survives under a separated-net hypothesis.
Since the entropy argument is label-aware, it could be adapted to random-center or measure-theoretic versions of the skeleton problem where labels are preserved.

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

Short version: the endpoint lower bound is real, the proof works, and the paper deserves a serious referee. I read the induction closely and couldn't find a load-bearing gap.

What's actually new: Thornton proved all sub-endpoint exponents and constructed examples at the endpoint, but for k≥1 the matching lower bound was open. This paper provides it, with a clean labelled-Shearer step that keeps coordinate labels attached to projections, and a one-step balanced induction instead of an iterated exponent-improvement map. The construction section also corrects Thornton's digit set; the correction matters and is handled well.

The proof is self-contained: Lemma 2.1 proves the uniform-cover entropy inequality, Lemma 3.1 passes from cube vertices to cofactors without losing multiplicity, and Proposition 4.1 closes strong induction with the same constant at the endpoint. I checked the small-radius branch: the cell bound N_m ≤ h^n is exactly what makes the induction descend, and the overlap bound is the standard 3^n. The constants in (4.4) work. The digit construction in Lemma 5.1 also checks out—the cancellation argument is right.

The soft spots are minor. The paper is lattice-only; the small-radius branch genuinely needs the cell occupancy bound. That's a real limitation, acknowledged in §6, but it's not a flaw in the stated theorem. Also, in the proof of Lemma 2.2 there's a typo: it writes |C| ≥ √2 |T| ∑_L √p_L, but it should be √(2|T|) (or √2 √|T|) times the sum; the final bound is correct, so the slip doesn't affect the result. There's also a stray "Theorem 2.1" reference where it means Lemma 2.1. Nothing else.

Who benefits: anyone working on discrete cube skeletons, square centers, or discretized maximal operators. The lattice restriction may limit direct application, but the technique—the labelled uniform-cover inequality and the balanced induction—is the kind of thing that could be reused.

My call: send it to a serious referee. If the referee goes through the induction once, they'll likely accept. I'd cite it if I worked in this area.

Referee Report

0 major / 6 minor

Summary. The paper studies the minimum size of a finite lattice set B that contains a filled axis-parallel discrete k-skeleton S_k(x,r_x) about each point of an N-point set S of lattice centers. The main result (Theorem 1.1) is that F_{n,k}(N) is bounded between positive constants times N^{1-(n-k)/(2n^2)} for all N. The lower bound, previously open for k≥1, is proved via a labelled Shearer entropy inequality, a midpoint estimate for cube vertices, a cofactor estimate for projections onto (n−k)-coordinate faces, and a strong induction balancing large- and small-radius cases without a logarithmic loss. The upper bound is a corrected digit construction with the same exponent.

Significance. The result resolves the endpoint exponent for discrete cube skeleta, closing the gap between Thornton's sub-endpoint lower bounds and his construction. The proof is self-contained and transparent: Lemma 2.1 supplies a labelled uniform-cover entropy inequality, Lemma 3.1 converts vertex data into normal cofactors, and Proposition 4.1 closes the induction with a single constant at the endpoint. The digit construction in Section 5 gives a matching upper bound and is independently verified. The paper also states its methodological limitation for nonlattice centers (§6), which is correctly scoped. If the proof is valid, this is a substantial contribution to discrete geometry.

minor comments (6)

[§2 (proof of Lemma 2.2)] The sentence 'Apply Theorem 2.1 to the n coordinate-deletion sets' should refer to Lemma 2.1; there is no Theorem 2.1. The same mislabelling appears in the proof of Lemma 3.1: 'Consequently Theorem 2.1 gives' should be 'Lemma 2.1'.
[§4 (proof of Proposition 4.1)] The sentence 'Thus the pair A, S_L satisfies the hypotheses of Theorem 3.1' should refer to Lemma 3.1, not a Theorem.
[§5 (proof of Theorem 1.1)] In the final proof, 'Theorem 4.1 applies' and 'Theorem 5.2 supplies' are mislabelled; these are Proposition 4.1 and Proposition 5.2.
[§1 (Introduction)] The road-map paragraph refers to 'Theorems 2.1 to 2.3', 'Theorem 3.1', 'Theorem 4.1', and 'Theorems 5.1 and 5.2'. These are actually Lemmas 2.1–2.3, Lemma 3.1, Proposition 4.1, and Lemma 5.1/Proposition 5.2. Please correct the labels throughout for consistency.
[Full-text title] The title as rendered in the full text contains a typo: 'CUBE SKELET A' should likely read 'CUBE SKELETA' or 'CUBE SKELETA' as in the abstract.
[Abstract] The phrase 'Thornton proved every smaller exponent' is imprecise; it would be clearer to say 'Thornton proved the lower bound for every exponent smaller than the endpoint'.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the lower bound, entropy/cofactor estimates, induction, and construction are self-contained.

full rationale

The paper's central derivation does not reduce to its own inputs. Lemma 2.1 proves the labelled uniform-cover Shearer inequality directly; Lemmas 2.2 and 2.3 prove the axial midpoint bound and cube-vertex completion from elementary pair counts and an invertible linear map; Lemma 3.1 passes from vertices to cofactors via the completion set C and the same entropy inequality, with no appeal to the target bound. Proposition 4.1 closes by strong induction: the large-radius branch uses the cofactor estimate plus disjointness of parallel faces, while the small-radius branch uses the lattice occupancy bound N_m ≤ h^n, which is stated and is acknowledged in §6 as a limitation of the method for nonlattice centers. The induction hypothesis applies only to smaller center sets with the same constant, and the constants are chosen explicitly from (4.4) and (4.5). The upper-bound construction in Lemma 5.1 and Proposition 5.2 is a self-contained digit construction, explicitly a corrected version of Thornton's construction, and does not assume the theorem. External citations (Shearer, Keleti–Nagy–Shmerkin, Olivo–Shmerkin, Thornton) are used for context and comparison, not as load-bearing inputs, and there are no self-citations by the author. No fitted parameter is renamed as a prediction, and no uniqueness theorem is imported from prior work of the present author. The proof is therefore internally self-contained, with no circular step identified.

Assumptions & free parameters 1 free parameters · 2 assumptions · 0 invented entities

No data-fitted parameters are used; the auxiliary smallness parameter a affects only constants. No new entities or mediators are postulated. The principal unproved inputs are the standard entropy toolkit and the explicit lattice-domain assumption.

free parameters (1)

a (smallness parameter) = any real in (0,1) satisfying 2·3^n ρ^{1-β}≤1 with ρ=2^n a^n
Introduced in §4 (Eqs. 4.3–4.4) to make ρ small enough for the small-radius induction step; it affects only the constant c_{n,k}, not the exponent β. It is an auxiliary proof parameter rather than a fitted model quantity.

assumptions (2)

domain assumption Centers S and carrier B are finite subsets of Z^n and radii are positive integers (definition of F_{n,k}).
The theorem is stated for lattice sets. The small-radius branch of Proposition 4.1 requires that a cell of side h contain at most h^n centers (Eq. 4.8); the paper's concluding remark says nonlattice centers would require an additional separated-net assumption.
standard math Standard Shannon entropy facts and strong induction.
The proof uses the chain rule, conditioning-reduces-entropy, weighted AM-GM, and induction on N without proof; these are standard background mathematics.

how reviews work

0 comments

Cite this review

Pith. "Pith review of The Endpoint Cardinality of Discrete Cube Skeleta." pith.science (2026). https://pith.science/paper/ONGI6FMD

@misc{pith2026260715502,
  author       = {Pith},
  title        = {Pith review of: The Endpoint Cardinality of Discrete Cube Skeleta},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/ONGI6FMD}},
  note         = {Machine review of arXiv:2607.15502}
}

abstract

We determine the minimum order of a finite lattice set that contains a filled axis-parallel cube skeleton about every point of some $N$-point set of centers. For fixed integers $0\leq k<n$, the answer for $N$ centers is $N^{1-(n-k)/(2n^2)}$, up to constants depending on $n$ and $k$. Thornton proved every smaller exponent and gave a construction of this order; the endpoint lower bound was left open when $k\geq1$. Our proof combines a midpoint estimate, a labelled form of Shearer's projection inequality, and a strong induction that balances large and small radii without a dyadic pigeonhole loss. In particular, a lattice set containing a square boundary about each of $N$ centers has at least a constant times $N^{7/8}$ points.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

25 extracted references · 11 linked inside Pith

[1]

Acta Mathematica Hungarica , volume =

Thornton, Riley , title =. Acta Mathematica Hungarica , volume =. 2017 , doi =. 1502.02187 , archiveprefix =

arXiv 2017
[2]

Squares and Their Centers , journal =

Keleti, Tam. Squares and Their Centers , journal =. 2018 , doi =. 1408.1029 , archiveprefix =

arXiv 2018
[3]

Annales Academiae Scientiarum Fennicae Mathematica , volume =

Olivo, Andrea and Shmerkin, Pablo , title =. Annales Academiae Scientiarum Fennicae Mathematica , volume =. 2020 , doi =. 1807.05280 , archiveprefix =

arXiv 2020
[4]

Small Unions of Affine Subspaces and Skeletons via

Chang, Alan and Cs. Small Unions of Affine Subspaces and Skeletons via. Advances in Mathematics , volume =. 2018 , doi =. 1701.01405 , archiveprefix =

arXiv 2018
[5]

Small Union with Large Set of Centers , booktitle =

Keleti, Tam. Small Union with Large Set of Centers , booktitle =. 2017 , doi =. 1701.02762 , archiveprefix =

arXiv 2017
[6]

Hausdorff Dimension of Unions of Affine Subspaces and of

H. Hausdorff Dimension of Unions of Affine Subspaces and of. Journal of Fractal Geometry , volume =. 2019 , doi =. 1701.02299 , archiveprefix =

arXiv 2019
[7]

Chung, Fan R. K. and Graham, Ronald L. and Frankl, Peter and Shearer, James B. , title =. Journal of Combinatorial Theory, Series A , volume =. 1986 , doi =

1986
[8]

, title =

Karp, Richard M. , title =. Discrete Mathematics , volume =. 1978 , doi =

1978

Show all 25 references

[9]

de Bruijn, N. G. , title =. Proceedings of the Section of Sciences of the Koninklijke Nederlandse Akademie van Wetenschappen te Amsterdam , volume =
[10]

and Thomas, Joy A

Cover, Thomas M. and Thomas, Joy A. , title =
[11]

Hedayat, A. S. and Sloane, N. J. A. and Stufken, John , title =. 1999 , doi =

1999
[12]

Bailey, R. A. and Cameron, Peter J. and Kinyon, Michael and Praeger, Cheryl E. , title =. Designs, Codes and Cryptography , volume =. 2022 , doi =. 2010.16338 , archiveprefix =

2022 arXiv
[13]

Falconer, Kenneth , title =
[14]

Mathematical Proceedings of the Cambridge Philosophical Society , volume =

Tricot, Claude , title =. Mathematical Proceedings of the Cambridge Philosophical Society , volume =. 1982 , doi =

1982
[15]

and Larrabee, Tracy L

Knuth, Donald E. and Larrabee, Tracy L. and Roberts, Paul M. , title =. 1989 , isbn =

1989
[16]

and Whitney, Hassler , title =

Loomis, Lynn H. and Whitney, Hassler , title =. Bulletin of the American Mathematical Society , volume =. 1949 , doi =

1949
[17]

Discrete Analysis , number =

Ellis, David and Friedgut, Ehud and Kindler, Guy and Yehudayoff, Amir , title =. Discrete Analysis , number =. 2016 , doi =. 1510.00258 , archiveprefix =

2016 arXiv
[18]

2013 , eprint =

Christ, Michael , title =. 2013 , eprint =

2013
[19]

Sobel, Milton and Uppuluri, V. R. R. , title =. The Annals of Mathematical Statistics , volume =. 1972 , doi =

1972
[20]

Kwerel, S. M. , title =. Journal of Applied Probability , volume =. 1975 , doi =

1975
[21]

2015 , eprint =

Christ, Michael and Demmel, James and Knight, Nicholas and Scanlon, Thomas and Yelick, Katherine , title =. 2015 , eprint =

2015
[22]

Geometric and Functional Analysis , volume =

Bennett, Jonathan and Carbery, Anthony and Christ, Michael and Tao, Terence , title =. Geometric and Functional Analysis , volume =. 2008 , doi =. math/0505065 , archiveprefix =

2008 arXiv
[23]

Mathematika , volume =

Yu, Han , title =. Mathematika , volume =. 2021 , doi =. 1711.06544 , archiveprefix =

2021 arXiv
[24]

2020 , eprint =

Cowen-Breen, Charlie and Karangozishvili, Elene and Varadarajan, Narmada and Wang, Thomas , title =. 2020 , eprint =

2020
[25]

, title =

Fiedler, Jacob B. , title =. The Journal of Geometric Analysis , volume =. 2026 , doi =. 2508.18257 , archiveprefix =

2026 arXiv

Pith tools

Reviewed August 1, 2026 · model on record in the stance chip above.

[1] [1]

Acta Mathematica Hungarica , volume =

Thornton, Riley , title =. Acta Mathematica Hungarica , volume =. 2017 , doi =. 1502.02187 , archiveprefix =

arXiv 2017

[2] [2]

Squares and Their Centers , journal =

Keleti, Tam. Squares and Their Centers , journal =. 2018 , doi =. 1408.1029 , archiveprefix =

arXiv 2018

[3] [3]

Annales Academiae Scientiarum Fennicae Mathematica , volume =

Olivo, Andrea and Shmerkin, Pablo , title =. Annales Academiae Scientiarum Fennicae Mathematica , volume =. 2020 , doi =. 1807.05280 , archiveprefix =

arXiv 2020

[4] [4]

Small Unions of Affine Subspaces and Skeletons via

Chang, Alan and Cs. Small Unions of Affine Subspaces and Skeletons via. Advances in Mathematics , volume =. 2018 , doi =. 1701.01405 , archiveprefix =

arXiv 2018

[5] [5]

Small Union with Large Set of Centers , booktitle =

Keleti, Tam. Small Union with Large Set of Centers , booktitle =. 2017 , doi =. 1701.02762 , archiveprefix =

arXiv 2017

[6] [6]

Hausdorff Dimension of Unions of Affine Subspaces and of

H. Hausdorff Dimension of Unions of Affine Subspaces and of. Journal of Fractal Geometry , volume =. 2019 , doi =. 1701.02299 , archiveprefix =

arXiv 2019

[7] [7]

Chung, Fan R. K. and Graham, Ronald L. and Frankl, Peter and Shearer, James B. , title =. Journal of Combinatorial Theory, Series A , volume =. 1986 , doi =

1986

[8] [8]

, title =

Karp, Richard M. , title =. Discrete Mathematics , volume =. 1978 , doi =

1978