pith. sign in

arxiv: 2606.01347 · v1 · pith:VRHLHMC7new · submitted 2026-05-31 · 🧮 math.NT · math.FA· math.MG

A Sharp Reverse Minkowski Inequality for the Gaussian Mass of Integral Unimodular Lattices Through Rank 32

Scott Duke Kominers This is my paper

Pith reviewed 2026-06-28 16:29 UTC · model grok-4.3

classification 🧮 math.NT math.FAmath.MG MSC 11H0611F03
keywords integral unimodular latticesGaussian masstheta seriesreverse Minkowski inequalitymodular formsADE root systemsshadow positivity
0
0 comments X

The pith

The integer lattice Z^n maximizes Gaussian mass among integral unimodular lattices of rank at most 32, with equality only for that lattice.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper proves that among integral unimodular lattices in Euclidean space of dimension at most 32, the standard integer lattice Z^n has the largest Gaussian mass for every positive scaling parameter t. Equality occurs only when the lattice is isomorphic to Z^n. The same strict maximization holds for every even unimodular lattice in dimension 40. The argument expresses every such lattice's theta series as a polynomial in the modular function u = Δ8/ϑ3^8 whose coefficients are fixed by the number of norm-1 vectors, the ADE root systems present, and positivity properties of the shadow. This representation replaces the need to classify lattices in each dimension.

Core claim

We prove the sharp inequality that the Gaussian mass Θ_L(t) of any integral unimodular lattice L of rank n ≤ 32 is at most that of Z^n, with equality only when L ≅ Z^n. For even unimodular lattices of rank 40 the inequality is strict. Integral unimodular theta series are parametrized as polynomials in u = Δ8/ϑ3^8 ∈ (0,1/64] whose coefficients are controlled solely by norm-1 splitting, ADE root counts, and shadow positivity.

What carries the argument

The modular function u = Δ8/ϑ3^8 ∈ (0,1/64] that expresses every integral unimodular theta series as a polynomial whose coefficients are fixed by norm-1 splitting, ADE root counts, and shadow positivity.

If this is right

  • Z^n is the unique maximizer of Gaussian mass among integral unimodular lattices in every rank up to 32.
  • The inequality is strict for all even unimodular lattices in rank 40.
  • The parametrization by u works uniformly across ranks without requiring explicit lattice classification.
  • Equality in the inequality is attained precisely when the lattice is the standard integer lattice.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same coefficient controls on u might allow verification of the maximization conjecture in selected higher ranks where the parametrization still holds.
  • The emphasis on norm-1 vectors and ADE roots suggests these combinatorial features could bound other extremal functions on lattices, such as minimal vectors or packing density.
  • If the shadow positivity constraint can be relaxed or replaced, the method might extend to certain non-unimodular or non-integral families.

Load-bearing premise

Every integral unimodular theta series can be written as a polynomial in u=Δ8/ϑ3^8 whose coefficients are controlled solely by norm-1 splitting, ADE root counts, and shadow positivity.

What would settle it

An integral unimodular lattice of rank at most 32 whose Gaussian mass exceeds that of Z^n for some t > 0, or one achieving equality but not isomorphic to Z^n, or whose theta series cannot be expressed as such a polynomial with the stated coefficient controls.

read the original abstract

The integer lattice $\mathbb{Z}^n$ is conjectured to maximize the Gaussian mass $\Theta_L(t)=\sum_{x\in L}e^{-t\|x\|^2}$ over the set of stable lattices in $\mathbb{R}^n$, for every $t>0$. We prove this sharp inequality for every integral unimodular lattice $L$ of rank $n\leq 32$, with equality only at $L\cong\mathbb{Z}^n$, and furthermore obtain the strict inequality for every even unimodular lattice of rank $40$. The proof does not use the classification of unimodular lattices in these ranks; rather, it parametrizes integral unimodular theta series as polynomials in the modular function $u=\Delta_8/\vartheta_3^8\in(0,1/64]$, with the few coefficients that arise controlled by norm-$1$ splitting, ADE root counts, and shadow positivity.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 1 minor

Summary. The paper claims to prove a sharp reverse Minkowski inequality asserting that the Gaussian mass Θ_L(t) = ∑_{x∈L} e^{-t‖x‖²} is maximized by the integer lattice Z^n among all integral unimodular lattices L in rank n ≤ 32, with equality only when L ≅ Z^n; it further claims the strict inequality holds for every even unimodular lattice of rank 40. The argument parametrizes every integral unimodular theta series as a polynomial in the modular function u = Δ_8 / ϑ_3^8 ∈ (0, 1/64], with the arising coefficients controlled exclusively by the number of norm-1 vectors, ADE root-system counts, and shadow positivity, thereby avoiding any appeal to the classification of unimodular lattices.

Significance. If the central parametrization and coefficient bounds hold, the result supplies a classification-free verification of the conjectured maximality of Z^n for Gaussian mass in these ranks, together with an explicit modular-form reduction that may extend to higher ranks. The explicit use of standard lattice invariants (norm-1 splitting, ADE counts, shadow positivity) to bound coefficients is a concrete strength that renders the comparison to Z^n falsifiable in principle.

major comments (2)
  1. [abstract] Abstract (proof-strategy paragraph): the assertion that every integral unimodular theta series is a polynomial in u whose coefficients are controlled solely by norm-1 splitting, ADE root counts, and shadow positivity is load-bearing for the entire comparison; the manuscript must supply an explicit theorem or lemma giving the admissible coefficient ranges (or at least the sign and magnitude bounds) for each rank ≤ 32, otherwise the reduction to the Z^n case cannot be verified.
  2. [abstract] The rank-40 claim for even unimodular lattices is stated separately; the same polynomial parametrization and coefficient controls must be shown to apply verbatim in that case, or the strict inequality rests on an unexamined extension of the argument used for n ≤ 32.
minor comments (1)
  1. [abstract] The interval (0, 1/64] for u should be justified by a short reference or computation showing that this is the precise range attained by integral unimodular lattices.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the careful reading and for highlighting the need for greater explicitness in the coefficient controls. We address each major comment below and will incorporate the requested clarifications.

read point-by-point responses

  1. Referee: [abstract] Abstract (proof-strategy paragraph): the assertion that every integral unimodular theta series is a polynomial in u whose coefficients are controlled solely by norm-1 splitting, ADE root counts, and shadow positivity is load-bearing for the entire comparison; the manuscript must supply an explicit theorem or lemma giving the admissible coefficient ranges (or at least the sign and magnitude bounds) for each rank ≤ 32, otherwise the reduction to the Z^n case cannot be verified.

    Authors: We agree that an explicit statement of the admissible coefficient ranges is required for independent verification. Section 3 of the manuscript already derives the polynomial form and obtains the bounds from norm-1 splitting, ADE root counts, and shadow positivity, but these derivations are distributed across several propositions. We will add a single consolidated Lemma 3.1 that states the precise sign and magnitude bounds on each coefficient for every n ≤ 32. This lemma will make the comparison with the Z^n case fully checkable from the listed invariants alone. revision: yes

  2. Referee: [abstract] The rank-40 claim for even unimodular lattices is stated separately; the same polynomial parametrization and coefficient controls must be shown to apply verbatim in that case, or the strict inequality rests on an unexamined extension of the argument used for n ≤ 32.

    Authors: Even unimodular lattices of rank 40 are a subclass of integral unimodular lattices, so the same parametrization of the theta series as a polynomial in u applies directly. The coefficient bounds derived from shadow positivity and the other invariants remain identical; no new controls are needed. We will insert a short remark immediately after the statement of the main theorem for n ≤ 32, together with a sentence in the introduction, confirming that the identical bounds yield the strict inequality for all even unimodular lattices in rank 40. revision: yes

Circularity Check

0 steps flagged

No significant circularity identified

full rationale

The derivation parametrizes integral unimodular theta series as polynomials in the modular function u=Δ8/ϑ3^8 with coefficients controlled by independent invariants (norm-1 splitting, ADE root counts, shadow positivity). These controls are standard in the theory of lattice theta series and modular forms; they are not defined from the target Gaussian mass inequality, nor obtained by fitting to the same data, nor justified solely by self-citation. The strategy is explicitly classification-free and relies on external modular-function identities and positivity conditions that stand outside the claimed inequality, rendering the argument self-contained.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The proof rests on the assumption that theta series admit the stated polynomial parametrization and that the listed controls suffice to bound coefficients; no free parameters or invented entities are introduced in the abstract.

axioms (1)
  • domain assumption Integral unimodular theta series are polynomials in u=Δ8/ϑ3^8 with coefficients controlled by norm-1 splitting, ADE root counts, and shadow positivity
    Explicitly invoked in the abstract as the basis of the proof that avoids classification.

pith-pipeline@v0.9.1-grok · 5695 in / 1261 out tokens · 29903 ms · 2026-06-28T16:29:03.504299+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Majorization and Gaussian-Mass Maximality for Construction-A Lattices from Binary Self-Dual Codes

    math.NT 2026-06 unverdicted novelty 7.0

    Proves the Regev-Stephens-Davidowitz conjecture on Gaussian mass maximality of the integer lattice for unimodular Construction-A lattices from binary self-dual codes via majorization of half-weight distributions and J...

Reference graph

Works this paper leans on

29 extracted references · 3 canonical work pages · cited by 1 Pith paper · 2 internal anchors

  1. [1]

    Ann.296(1993), no

    Wojciech Banaszczyk,New bounds in some transference theorems in the geometry of numbers, Math. Ann.296(1993), no. 4, 625–635

    1993

  2. [2]

    Jean-Claude Belfiore and Patrick Sol´ e,Unimodular lattices for the Gaussian wiretap channel, Proceedings of the IEEE Information Theory Workshop, 2010, pp. 1–5

    2010

  3. [3]

    Bollauf and Hsuan-Yin Lin, On the maximum flatness factor over unimodular lattices, preprint, arXiv:2403.16932, 2025

    Maiara F. Bollauf and Hsuan-Yin Lin,On the maximum flatness factor over unimodular lattices, preprint, arXiv:2403.16932, 2024

    work page arXiv 2024

  4. [4]

    Bollauf, Hsuan-Yin Lin, and Øyvind Ytrehus,Formally unimodular packings for the Gaussian wiretap channel, IEEE Trans

    Maiara F. Bollauf, Hsuan-Yin Lin, and Øyvind Ytrehus,Formally unimodular packings for the Gaussian wiretap channel, IEEE Trans. Inform. Theory69(2023), no. 12, 7755–7776

    2023

  5. [5]

    Bill Casselman,Stability of lattices and the partition of arithmetic quotients, Asian J. Math. 8(2004), no. 4, 607–638

    2004

  6. [6]

    Miller, Danylo Radchenko, and Maryna Viazovska, The sphere packing problem in dimension24, Ann

    Henry Cohn, Abhinav Kumar, Stephen D. Miller, Danylo Radchenko, and Maryna Viazovska, The sphere packing problem in dimension24, Ann. of Math.185(2017), no. 3, 1017–1033

    2017

  7. [7]

    of Math.196(2022), no

    ,Universal optimality of theE 8 and Leech lattices and interpolation formulas, Ann. of Math.196(2022), no. 3, 983–1082

    2022

  8. [8]

    Comp.91(2022), no

    Henry Cohn and Nicholas Triantafillou,Dual linear programming bounds for sphere packing via modular forms, Math. Comp.91(2022), no. 333, 491–508

    2022

  9. [9]

    Conway and Neil J

    John H. Conway and Neil J. A. Sloane,Sphere Packings, Lattices and Groups, 3rd ed., Grundlehren der Mathematischen Wissenschaften, vol. 290, Springer, New York, 1999

    1999

  10. [10]

    Daniel Dadush and Oded Regev,Towards strong reverse Minkowski-type inequalities for lattices, Proceedings of the 57th IEEE Symposium on Foundations of Computer Science (FOCS), 2016, pp. 447–456

    2016

  11. [11]

    Wolfgang Ebeling,Lattices and Codes: A Course Partially Based on Lectures by Friedrich Hirzebruch, 3rd ed., Springer Spektrum, Wiesbaden, Germany, 2013

    2013

  12. [12]

    Yael Eisenberg, Oded Regev, and Noah Stephens-Davidowitz,A tight reverse Minkowski inequality for the Epstein zeta function, Proc. Amer. Math. Soc. (2026), to appear

    2026

  13. [13]

    Elkies,Lattices and codes with long shadows, Math

    Noam D. Elkies,Lattices and codes with long shadows, Math. Res. Lett.2(1995), no. 5, 643–651

    1995

  14. [14]

    Anne-Maria Ernvall-Hyt¨ onen,On a conjecture by Belfiore and Sol´ e on some lattices, IEEE Trans. Inform. Theory58(2012), no. 9, 5950–5955

    2012

  15. [15]

    Grayson,Reduction theory using semistability, Comment

    Daniel R. Grayson,Reduction theory using semistability, Comment. Math. Helv.59(1984), no. 4, 600–634

    1984

  16. [16]

    Danske Vi- densk

    Erich Hecke,Analytische Arithmetik der positiven quadratischen Formen, Kgl. Danske Vi- densk. Selsk. Math.-fys. Medd.XVII(1940), no. 12, 1–134, Reprinted inMathematische Werke, Vandenhoeck & Ruprecht, G¨ ottingen, 1959, pp. 789–918

    1940

  17. [17]

    Arne Heimendahl, Aurelio Marafioti, Antonia Thiemeyer, Frank Vallentin, and Marc Chris- tian Zimmermann,Critical even unimodular lattices in the Gaussian core model, Int. Math. Res. Not. IMRN (2023), no. 6, 5352–5396

    2023

  18. [18]

    King,A mass formula for unimodular lattices with no roots, Math

    Oliver D. King,A mass formula for unimodular lattices with no roots, Math. Comp.72 (2003), no. 242, 839–863

    2003

  19. [19]

    Scott Duke Kominers,Equality in a reverse Minkowski shell bound for integral lattices via spherical designs, preprint, arXiv:2605.25126, 2026

    work page internal anchor Pith review Pith/arXiv arXiv 2026

  20. [20]

    ,Saturation and no-go theorems for scalar Poisson certificates of Gaussian mass maximality, preprint, arXiv:2605.26803, 2026

    work page internal anchor Pith review Pith/arXiv arXiv 2026

  21. [21]

    Julia Pinchak and B. A. Sethuraman,The Belfiore–Sol´ e conjecture and a certain technique for verifying it for a given lattice, Proceedings of the Information Theory and Applications Workshop, 2014, pp. 486–488

    2014

  22. [22]

    Rains and Neil J

    Eric M. Rains and Neil J. A. Sloane,The shadow theory of modular and unimodular lattices, J. Number Theory73(1998), no. 2, 359–389

    1998

  23. [23]

    Oded Regev and Noah Stephens-Davidowitz,A reverse Minkowski theorem, Ann. of Math. 199(2024), no. 1, 1–49

    2024

  24. [24]

    Number Theory279(2026), 256–266

    ,A simple proof of a reverse Minkowski theorem for integral lattices, J. Number Theory279(2026), 256–266

    2026

  25. [25]

    W. P. Reinhardt and P. L. Walker,Theta functions, NIST Digital Library of Mathematical Functions, Release 1.2.6 (2026-03-15),https://dlmf.nist.gov/

    2026

  26. [26]

    Math.165(2006), no

    Peter Sarnak and Andreas Str¨ ombergsson,Minima of Epstein’s zeta function and heights of flat tori, Invent. Math.165(2006), no. 1, 115–151. REVERSE MINKOWSKI FOR GAUSSIAN MASS OF INTEGRAL UNIMODULAR LATTICES15

    2006

  27. [27]

    Ulrich Stuhler,Eine Bemerkung zur Reduktionstheorie quadratischer Formen, Arch. Math. (Basel)27(1976), no. 6, 604–610

    1976

  28. [28]

    of Math.185(2017), no

    Maryna Viazovska,The sphere packing problem in dimension8, Ann. of Math.185(2017), no. 3, 991–1015

    2017

  29. [29]

    E. T. Whittaker and G. N. Watson,A Course of Modern Analysis, 4th ed., Cambridge University Press, Cambridge, 1996. Harvard Business School; Department of Economics and Center of Mathematical Sciences and Applications, Harvard University; and a16z crypto Email address:kominers@fas.harvard.edu

    1996