Sharp refined-direction Kakeya estimates in finite Heisenberg groups

Andrea Pinamonti; Boqing Xue; Dung The Tran; Thang Pham

Sharp refined-direction Kakeya estimates in finite Heisenberg groups

T0 review · 0 major / 4 minor · reviewed 2026-08-27 · deepseek-v4-flash

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

Pith's one-line read In finite Heisenberg groups, the number of λ-rich refined directions is at most q^{2n-1}|E|/λ^{2n}, and this exponent is sharp.

desk verdict The n≥2 refined-direction Kakeya problem in finite Heisenberg groups is solved except for a single disclosed log factor; the horizontal-multiplicity method is a genuine new tool and the proof holds up. read the letter →

arxiv 2608.14059 v1 pith:JA2FFLBB submitted 2026-08-14 math.CA math.COmath.NT

Thang Pham , Andrea Pinamonti , Dung The Tran , Boqing Xue This is my paper · ORCID

classification math.CAmath.COmath.NT MSC 42B2505B2551E20

keywords finiteHeisenberggroupKakeyamaximaloperatorrefineddirectionpolynomialmethodofmultiplicitiesaffinesymplecticFurstenbergsetsfields

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 proves a sharp 'rich-direction' estimate for horizontal lines in finite Heisenberg groups over odd prime-power fields: for any set E and any threshold λ, the number of refined directions that contain a horizontal line with at least λ points of E is at most a constant times $q^{{2n-1}}$|E| $λ^{{-2n}}$. This is the natural Kakeya-type question when one records not only the spatial direction of a horizontal line but also its central slope. The authors also determine, for every ℓ^u→ℓ^v pair, the exact power of q in the corresponding maximal-operator estimate, given by the maximum of four simple affine expressions. The value is achieved by a pure power away from the critical diagonal (u,v)=(2n,2n), where only a (1+log q)^{1/(2n)} loss is known. These bounds matter because they give the sharp finite-field analogue of Kakeya compression for a genuinely higher-moment operator, with consequences for Heisenberg Furstenberg-type sets.

What carries the argument

The load-bearing device is horizontal polynomial multiplicity. For each point p=(z,t), the space of horizontal directions through p is H_p={(ξ,σ(z,ξ)): ξ∈$F_q^{{2n}}$}, an r=2n dimensional subspace of the ambient r+1 variables. Requiring that the Hasse–Taylor components of degree <m vanish on H_p imposes only binom(m+r-1,r)=O_n(m^r) linear conditions per point, rather than the O_n($m^{{r+1}}$) conditions of ordinary multiplicity. This dimension count lets the polynomial-method interpolation lemma produce a nonzero polynomial P of degree about |E|^{1/(r+1)} $m^{{r/(r+1)}}$ that vanishes with multiplicity m at every point of E along every horizontal direction. On a normalized line L_w(ρ) containing λ points of E, the univariate restriction has at least mλ zeros counted with multiplicity; choosing m so that mλ > deg P forces the restriction to vanish identically, and the top-homogeneous part Γ_P(w) then vanishes on all rich normalized directions, so the standard polynomial zero-counting bound limits their number. A second mechanism, the affine symplectic group acting transitively on refined directions, is used in a probabilistic covering step to pass from a sparse family of rich directions to a dense one, which is what removes an unwanted |E|^{1/(r+1)} term and yields the global bound.

What would settle it

For n=2, take q=5,7,11 and a random set E with |E|=$q^{2}$; for λ=⌊q/2⌋, compute |Ω_λ(E)| and the ratio |Ω_λ(E)| / ($q^{3}$|E|/$λ^{4}$). If this ratio grows with q, then Theorem 1.1 is false. The paper's point-mass and central-slice examples should be used as controls: they must keep the ratio bounded.

Watch

Extended reading notes

Core claim

The central discovery is that the refined-direction Kakeya maximal operator on H_n(F_q) satisfies |{ϑ : M^rd 1_E(ϑ) ≥ λ}| ≲_n $q^{{2n-1}}$|E| $λ^{{-2n}}$, with the exponent 2n-1 shown sharp by a point mass, and that the complete exponent diagram is A_n^rd(u,v) = max{(2n-1)/v, 1-1/u, 2n/v - 1/u, 1 + 2n/v - (2n+1)/u}. Each of the four terms is forced by a separate extremal configuration: a point, a single horizontal line, a small central-slice set meeting every affine hyperplane, and the constant function. The level-set estimate itself is logarithm-free; the logarithmic factor enters only when summing ordered values at the critical diagonal, leaving the pure power (2n-1)/(2n) as the known infimum there. In the n=1 case the same exponent diagram was previously known, but the present proof does not extend the Fourier argument: for n≥2 the critical diagonal moves to $ℓ^{{2n}}$, and the mechanism is a new higher-moment polynomial method.

Load-bearing premise

The argument depends on the assumption that demanding m-fold vanishing in the horizontal directions through a point costs only about $m^{{2n}}$ equations per point, not $m^{{2n+1}}$; if the cheaper count failed, the polynomial found would be too large to force the contradiction.

Editorial extensions

If this is right

For every set E and threshold λ, the number of refined directions carrying at least λ points of E is at most C_n q^{2n-1}|E|/λ^{2n}, and the power 2n-1 is optimal: a single point already creates about q^{2n-1} rich directions at λ=1.
A finite Heisenberg Furstenberg bound follows: if a set Ω of refined directions, with |Ω|=δ q^{2n}, is such that each direction admits a horizontal line containing λ points of E, then |E| is at least a constant times δ λ^{2n} q.
The complete mixed-norm exponent formula A_n^rd(u,v)=max{(2n-1)/v, 1-1/u, 2n/v - 1/u, 1 + 2n/v - (2n+1)/u} holds, with pure-power strong estimates at every pair except the critical diagonal (2n,2n), where only the logarithmic-factor estimate is proved.
A family of horizontal lines with pairwise distinct refined directions cannot overlap too heavily: the union of such lines has size about |L| q, linear in the number of lines.
The n=1 Fourier-based proof cannot be extended: for n≥2 the critical diagonal is ℓ^{2n}, so the refined-direction problem requires the new higher-moment polynomial mechanism.

Reading between the lines

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

If the logarithmic factor at the critical diagonal is removable, the natural route would be a rearrangement-free argument that avoids summing ordered values; testing whether the strong ℓ^{2n} estimate holds for specially designed functions with flat maximal profiles could indicate whether the loss is essential.
The same horizontal-multiplicity mechanism may extend to other two-step nilpotent groups over finite fields, where the dimension count for horizontal multiplicity would determine whether the sharp rich-direction exponent remains (2n-1)/2n.
One could numerically probe the sharpness examples for small q and n=2: computing |Ω_λ(E)| for point masses, single lines, central-slice sets, and constant functions should track the four predicted powers, and any deviation would expose a hidden constant issue.
The paper leaves open whether the infimum at (2n,2n) is attained; a construction of functions whose ordered-value sums genuinely grow like (log q)^{1/(2n)} over the level-set bound would settle that the logarithmic factor is necessary.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Request a human review

A listed scientist reviews the paper for a fee and the review publishes here regardless of verdict. See the reviewers or get listed.

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

Dear colleague,

Bottom line: this is a serious paper and should be refereed. It proves sharp q-exponents for the refined-direction maximal operator on finite Heisenberg groups for all n≥2, determining the full ℓ^u→ℓ^v diagram except for one explicitly disclosed endpoint. The critical diagonal moves from ℓ^2 (n=1) to ℓ^{2n}, and the paper explains why the earlier Fourier method cannot reach it. That alone is a real advance.

The main new mechanism is horizontal multiplicity: requiring Hasse components to vanish only on the r-dimensional horizontal subspace at each point, instead of full ambient multiplicity, cuts the condition count from O(m^{r+1}) to O(m^r) per point. The interpolation lemma and the degree-versus-zeros contradiction in Proposition 4.7 check out; the symplectic covering lemma is clean, and the four lower-bound examples really are the four affine functionals in the exponent formula. I cannot find a load-bearing gap. The self-citation to [10] is legitimate: that paper solves n=1, and this one explicitly proves that its method loses a q^{n/2} factor in higher n.

The soft spot is the one the authors themselves flag: at the critical diagonal (u,v)=(2n,2n), they get q^{(2n-1)/(2n)}(1+\log q)^{1/(2n)} and show the pure power is the infimum, but whether the log can be removed remains open. That is a limitation, not a contradiction; every other pair attains the pure power. A second, milder concern is bookkeeping: the argument spans a lot of moving parts—multiplicities, covering, weighted level sets, interpolation—and it is not machine-checked. The reader's confidence of MODERATE is about right; I would not bet the house on every constant, but the architecture is coherent and the omitted checks are routine.

I agree with the stress-test note. For a paper of this level, the proof is unusually transparent about its one unresolved point. This deserves a full referee report, not a desk reject. I would bring it to reading group.

Referee Report

0 major / 4 minor

Summary. This paper studies the refined-direction Kakeya maximal operator on finite Heisenberg groups H_n(F_q) for n≥2, where a refined direction records both the spatial direction of a horizontal line and its central slope. The first main result (Theorem 1.1) is the sharp rich-direction estimate |Ω_λ(E)| ≲_n q^{2n-1}|E|λ^{-2n} for arbitrary E⊂H_n(F_q) and λ>0, with the q-exponent shown sharp by a point-mass example. A weighted version (Theorem 1.2) gives the level-set estimate |{ϑ: M^rd F(ϑ)≥λ}| ≲ q^{2n-1}λ^{-2n}∥F∥_{2n}^{2n}. The second main result (Theorem 1.6) determines, for every 1≤u,v≤∞, the optimal power A_n^rd(u,v) as the maximum of four affine functionals, with pure-power estimates at every exponent pair except (u,v)=(2n,2n), where a (1+log q)^{1/(2n)} factor remains and the plain power is shown to be the infimum. The proof combines a horizontal-multiplicity polynomial method, a probabilistic covering argument using the affine symplectic group, and real/Riesz–Thorin interpolation.

Significance. If correct, this resolves the refined-direction Kakeya problem for finite Heisenberg groups in the full mixed-norm range, going beyond the Fourier-analytic n=1 case of [10]. The exponent diagram is exact and parameter-free, with four explicit test functions forcing the four affine functionals. The technical core is a new horizontal-multiplicity notion imposed only on the r-dimensional horizontal subspaces, which reduces the condition count from O(m^{r+1}) to O(m^r) per point; I checked Lemma 4.3 and the stress-test concern about the dimension count does not land: the count of binom(m+r-1,r) conditions per point is correct, and Lemma 4.4 consequently gives the stated degree bound. The covering argument (Lemma 5.2) using the affine symplectic group is elegant and reduces the global estimate to the affine-chart estimate. The proof is self-contained, with no fitted parameters, and the sharp exponents are falsifiable predictions. The one open point, the pure-power ℓ^{2n}→ℓ^{2n} estimate, is explicitly disclosed with the best known logarithmic factor, which is a stated limitation rather than an internal inconsistency.

minor comments (4)

[Section 7, Proposition 7.2] The inequality ν_cov(p)^r ≤ 2^r∑_j ν(g_j^{-1}(p))^r relies on the inner sum being either 0 or at least 1 because ν is integer-valued; stating this explicitly would help the reader follow the ceiling argument.
[Section 9, proof of Theorem 1.6] The letter P is reused for the critical exponent pair (1/r,1/r) after earlier denoting a polynomial; although the context is clear, a different symbol would avoid ambiguity.
[Section 4, Lemma 4.4] The absorption of the ceiling into the constant C_int is terse; one additional sentence showing deg P ≤ 2C_deg when N=m=1 would improve readability.
[Section 5, Lemma 5.2] The condition q^r≥4 is used in the final inequality; it holds for odd prime powers q and r=2n≥4, and noting this would be helpful.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the sharp exponents are derived from polynomial-method upper bounds and explicit test-function lower bounds; the only logarithmic-factor gap is disclosed as an open limitation, not hidden as a prediction.

full rationale

The derivation chain is self-contained. Theorem 1.1 is proved by a horizontal-multiplicity polynomial construction (Lemma 4.4), a degree-versus-multiplicity contradiction (Proposition 4.7), and an affine-symplectic covering argument (Lemma 5.2). The dimension count in Lemma 4.3 is a direct coefficient count: because the horizontal direction space H_p is the graph of a linear functional, each Hasse component restricted to H_p is a homogeneous polynomial in r variables, so multiplicity at least m imposes at most binom(m+r-1, r) conditions per point. This is neither assumed nor imported: it is established inside the paper. The weighted level-set estimate (Proposition 7.2) follows by the same internally proved mechanism. The critical diagonal bound (Theorem 8.1) is an honest upper estimate with the factor (1+log q)^{1/r}, and Remark 9.3 explicitly states that pure-power attainment at (2n,2n) is open; the equality A_n^rd(r,r)=(r-1)/r identifies the infimum, not an attained minimum. All lower bounds come from four explicit test functions — a point mass, one horizontal line, a small central-slice set, and the constant function — whose ratios are computed directly in Proposition 3.1, so the four affine functionals in the exponent formula are forced by examples, not by the upper-bound machinery. The self-citation to [10] is used for the n=1 Fourier result, for context, and to explain why the n>=2 proof requires a different mechanism; the n>=2 estimates do not invoke [10] as a load-bearing premise. No fitted parameter is renamed as a prediction, no uniqueness theorem is imported from the authors' prior work, and no ansatz is smuggled via citation. The only explicitly disclosed limitation is the logarithmic factor on the critical diagonal, which is flagged rather than concealed. Accordingly, the circularity score is 0.

Assumptions & free parameters 0 free parameters · 4 assumptions · 0 invented entities

No empirical or fitted constants appear. The auxiliary proof constants C_int, C_gap, C_rich, C_dim, and C_mult depend only on n and are chosen existentially; they do not enter the sharp q-exponents. Horizontal multiplicity is a mathematical definition within the proof, not an invented physical or combinatorial entity.

assumptions (4)

standard math Standard finite-field polynomial lemmas, including Hasse derivatives and the distinction between formal polynomials and polynomial functions over F_q.
Used throughout Section 4 to define horizontal multiplicity and to justify coefficient-wise vanishing conditions over finite fields.
standard math Schwartz-Zippel bound: a nonzero polynomial of degree D in r variables over F_q has at most D q^{r-1} zeros.
Invoked in Lemma 4.8(ii) and applied in Proposition 4.7 and Proposition 7.1 to bound the number of rich directions.
standard math Polynomial dimension-counting: a homogeneous linear system with more variables than constraints over F_q has a nonzero solution.
Used in Lemma 4.4 and Proposition 7.1 to construct the nonzero interpolating polynomials with controlled degree.
standard math Riesz-Thorin and Marcinkiewicz interpolation for finite-dimensional linear operators on counting-measure spaces.
Used in Section 9 to convert level-set bounds and corner estimates into strong mixed-norm estimates across the exponent diagram.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Sharp refined-direction Kakeya estimates in finite Heisenberg groups." pith.science (2026). https://pith.science/paper/JA2FFLBB

@misc{pith2026260814059,
  author       = {Pith},
  title        = {Pith review of: Sharp refined-direction Kakeya estimates in finite Heisenberg groups},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/JA2FFLBB}},
  note         = {Machine review of arXiv:2608.14059}
}

abstract

Let $n\geq 2$ and let $q$ be an odd prime power. The first aim of this paper is to prove that, for every $E\subset \mathbb{H}_n(\mathbb{F}_q)$ and every $\lambda>0$, the following sharp rich-direction estimate holds \[ \left| \left\{ \vartheta\in D_n: M^{\mathrm{rd}}_{\mathbb{H}_n}\mathbf{1}_E(\vartheta)\geq\lambda \right\} \right| \lesssim_n q^{2n-1}|E|\lambda^{-2n}. \] The second aim is to determine, for every $1\leq u,v\leq\infty$, the sharp exponent of $q$ in the corresponding $\ell^u\to\ell^v$ estimate. More precisely, we prove that \[ A_n^{\mathrm{rd}}(u,v) = \max\left\{ \frac{2n-1}{v},\ 1-\frac1u,\ \frac{2n}{v}-\frac1u,\ 1+\frac{2n}{v}-\frac{2n+1}{u} \right\}. \] The proof combines the polynomial method with multiplicities and a probabilistic covering argument based on the action of the affine symplectic group.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

14 extracted references · 14 canonical work pages

[10]

T. Pham, A. Pinamonti, D. T. Tran, and B. Xue,Horizontal Kakeya maximal operators in finite Heisenberg groups: exact exponents and applications, arXiv preprint arXiv:2603.02111 (2026)

work page Pith review arXiv 2026
[1]

M. Dhar, Z. Dvir, and B. Lund,Simple proofs for Furstenberg sets over finite fields, Discrete Analysis (2021), Paper No. 22, 16 pp

work page 2021
[2]

M. Dhar, Z. Dvir, and B. Lund,Furstenberg sets in finite fields: explaining and improving the Ellenberg–Erman proof, Discrete & Computational Geometry,71(2) (2024), 327–357

work page 2024
[3]

Dvir,On the size of Kakeya sets in finite fields, Journal of the American Mathe- matical Society,22(4) (2009), 1093–1097

Z. Dvir,On the size of Kakeya sets in finite fields, Journal of the American Mathe- matical Society,22(4) (2009), 1093–1097. REFINED-DIRECTION KAKEYA ESTIMATES 37

work page 2009
[4]

Z. Dvir, S. Kopparty, S. Saraf, and M. Sudan,Extensions to the method of multiplicities, with applications to Kakeya sets and mergers, SIAM Journal on Computing,42(6) (2013), 2305–2328

work page 2013
[5]

J. S. Ellenberg and D. Erman,Furstenberg sets and Furstenberg schemes over finite fields, Algebra & Number Theory,10(7) (2016), 1415–1436

work page 2016
[6]

J. S. Ellenberg, R. Oberlin, and T. Tao,The Kakeya set and maximal conjectures for algebraic varieties over finite fields, Mathematika,56(1) (2010), 1–25

work page 2010
[7]

F¨ assler, A

K. F¨ assler, A. Pinamonti, and P. Wald,A Kakeya maximal inequality in the Heisenberg group, Annali della Scuola Normale Superiore di Pisa, Classe di Scienze,26(3) (2025), 1451–1474

work page 2025

Show all 14 references

[8]

Liu,On the dimension of Kakeya sets in the first Heisenberg group, Proceedings of the American Mathematical Society,150(8) (2022), 3445–3455

J. Liu,On the dimension of Kakeya sets in the first Heisenberg group, Proceedings of the American Mathematical Society,150(8) (2022), 3445–3455

2022
[9]

Mockenhaupt and T

G. Mockenhaupt and T. Tao,Restriction and Kakeya phenomena for finite fields, Duke Mathematical Journal,121(1) (2004), 35–74

2004
[11]

Salvatore,The Kakeya conjecture on local fields of positive characteristic, Mathe- matika,69(1) (2023), 1–16

A. Salvatore,The Kakeya conjecture on local fields of positive characteristic, Mathe- matika,69(1) (2023), 1–16

2023
[12]

J. T. Schwartz,Fast probabilistic algorithms for verification of polynomial identities, Journal of the ACM,27(4) (1980), 701–717

1980
[13]

Venieri,Heisenberg Hausdorff dimension of Besicovitch sets, Analysis and Geometry in Metric Spaces,2(1) (2014), 319–327

L. Venieri,Heisenberg Hausdorff dimension of Besicovitch sets, Analysis and Geometry in Metric Spaces,2(1) (2014), 319–327

2014
[14]

Zippel,Probabilistic algorithms for sparse polynomials, in E

R. Zippel,Probabilistic algorithms for sparse polynomials, in E. W. Ng (ed.),Symbolic and Algebraic Computation: EUROSAM 1979, Lecture Notes in Computer Science, vol. 72, Springer-Verlag, Berlin–Heidelberg (1979), 216–226. Institute of Mathematics and Interdisciplinary Science...

1979

Pith tools

Reviewed August 27, 2026 · model on record in the stance chip above.

[1] [10]

T. Pham, A. Pinamonti, D. T. Tran, and B. Xue,Horizontal Kakeya maximal operators in finite Heisenberg groups: exact exponents and applications, arXiv preprint arXiv:2603.02111 (2026)

work page Pith review arXiv 2026

[2] [1]

M. Dhar, Z. Dvir, and B. Lund,Simple proofs for Furstenberg sets over finite fields, Discrete Analysis (2021), Paper No. 22, 16 pp

work page 2021

[3] [2]

M. Dhar, Z. Dvir, and B. Lund,Furstenberg sets in finite fields: explaining and improving the Ellenberg–Erman proof, Discrete & Computational Geometry,71(2) (2024), 327–357

work page 2024

[4] [3]

Dvir,On the size of Kakeya sets in finite fields, Journal of the American Mathe- matical Society,22(4) (2009), 1093–1097

Z. Dvir,On the size of Kakeya sets in finite fields, Journal of the American Mathe- matical Society,22(4) (2009), 1093–1097. REFINED-DIRECTION KAKEYA ESTIMATES 37

work page 2009

[5] [4]

Z. Dvir, S. Kopparty, S. Saraf, and M. Sudan,Extensions to the method of multiplicities, with applications to Kakeya sets and mergers, SIAM Journal on Computing,42(6) (2013), 2305–2328

work page 2013

[6] [5]

J. S. Ellenberg and D. Erman,Furstenberg sets and Furstenberg schemes over finite fields, Algebra & Number Theory,10(7) (2016), 1415–1436

work page 2016

[7] [6]

J. S. Ellenberg, R. Oberlin, and T. Tao,The Kakeya set and maximal conjectures for algebraic varieties over finite fields, Mathematika,56(1) (2010), 1–25

work page 2010

[8] [7]

F¨ assler, A

K. F¨ assler, A. Pinamonti, and P. Wald,A Kakeya maximal inequality in the Heisenberg group, Annali della Scuola Normale Superiore di Pisa, Classe di Scienze,26(3) (2025), 1451–1474

work page 2025