betaFeedback

WorksMathematics

Hadamard configurations and sharp cubic-period bounds for stealth in dimensions 10, 11, and 12

Abstract

We construct explicit, nonlattice periodic point configurations in dimensions 1010, 1111 and 1212 with stealth center density 1/241/24 and unit-density stealth radius 2⋅24−1/n2\cdot24^{-1/n}. These values exceed those of duals of the current record lattice packings, without claiming to exceed all possible lattices. The construction uses the 2424 sign rows of an order-1212 Hadamard matrix and their antipodes, projected onto nn coordinates. For an arbitrary, not necessarily binary, equally weighted motif with marked period 2Zn2\mathbb{Z}^n, we prove a sharp 2n2n-point lower bound for a raw stealth radius at least one and classify equality by antipodal Hadamard configurations. We deduce bounded-complexity optimality, sharp optimality among binary motifs in dimensions 1010–1212, and local maximality under simultaneous perturbations of the motif and the marked period shape. A divisibility argument excludes ordinary equally weighted periodic formal duals. Statements are confined to their indicated classes; no global optimality among arbitrary-period configurations is asserted.

Introduction and main results

For a periodic point set, stealthiness is the radius of the largest open ball about the origin free of nonforward diffraction peaks. Cohen, Cohn, and Viazovska [1] recently established sharp global theorems in dimensions 8 and 24 and constructed a dimension-20 nonlattice point set outperforming all known lattices. The problem of whether the globally optimal stealth configuration must be a lattice remains open in many dimensions. The distinction between a periodic set and a single lattice is essential: a basis form factor can cancel diffraction shells.

Our constructions arise from classical Hadamard orthogonal arrays, but their interpretation as stealth configurations and the accompanying marked-cubic-period rigidity statements are the focus here. Let vnv_n denote the volume of the Euclidean unit ball. For a periodic set PP of point density ρ(P)\rho(P) and raw stealth radius K(P)K(P), define the scale-invariant stealth center density

Δst(P)=(K(P)/2)nρ(P).(1)\Delta_{\mathrm{st}}(P)=\frac{(K(P)/2)^n}{\rho(P)}. \tag*{(1)}

In the normalization of unit particle density this gives K=2Δst1/nK=2\Delta_{\mathrm{st}}^{1/n}. On lattices the quantity is the center density of the dual lattice; for general periodic configurations it need not be a sphere-packing density.

Theorem 1.1 (Explicit examples). For each n∈{10,11,12}n \in\{10,11,12\} there exists a periodic point configuration Pn⊂RnP_n \subset\mathbb{R}^n with 24 representatives modulo 2Zn2\mathbb{Z}^n such that

K(Pn)=1,ρ(Pn)=242n,Δst(Pn)=124.K(P_n)=1,\qquad\rho(P_n)=\frac{24}{2^n},\qquad\Delta_{\mathrm{st}}(P_n)=\frac{1}{24}.

The configurations are not translates of lattices. At unit particle density their stealth radii are 2⋅24−1/n2\cdot24^{-1/n}, exceeding the stealth radii of duals of the current record lattice packings in those dimensions.

Theorem 1.2 (Sharp cubic-torus cardinality and equality). Let n≥3n\geq3 and let P=X+2ZnP=X+2\mathbb{Z}^n, with NN distinct equally weighted representatives X⊂[0,2)nX\subset[0,2)^n. If K(P)≥1K(P)\geq1, then N≥2nN\geq2n. If N=2nN=2n, then, after a common translation, XX consists of the binary sign rows associated with HH and −H-H for a Hadamard matrix HH of order nn. Conversely, every such antipodal Hadamard array achieves K(P)=1K(P)=1.

Theorem 1.3 (Local maximality in a marked period model). Let n≥4n\geq4 and let C⊂{0,1}nC\subset\{0,1\}^n be a strength-three binary orthogonal array of NN distinct rows with N≡8(mod16)N\equiv8\pmod{16}. Then C+2ZnC+2\mathbb{Z}^n is locally maximal for Δst\Delta_{\mathrm{st}} against simultaneous sufficiently small changes in the positions of its NN labelled representatives and in its marked period matrix. Every sufficiently small nonsimilarity change of period shape strictly decreases Δst\Delta_{\mathrm{st}}. For n=12n=12 and N=24N=24, the equality case is locally isolated modulo translations, similarities, and point relabelling.

The marked period is not necessarily primitive: for our arrays the extra translation (1,…,1)(1,\ldots,1) is a period, and the primitive cell has 12, rather than 24, representatives. Neither Theorem 1.2 nor Theorem 1.3 is a global theorem over arbitrary period shapes or arbitrary supercells.

Diffraction and its normalization

Fix the Fourier kernel e−2πix⋅ξe^{-2\pi i x\cdot\xi}. If L⊂RnL\subset\mathbb{R}^n is a full-rank lattice and P=X+LP=X+L has NN distinct representatives modulo LL, then the support of its diffraction away from the forward peak consists precisely of

ξ∈L∗∖{0}withAX(ξ):=∑x∈Xe2πiξ⋅x≠0.\xi\in L^*\setminus\{0\}\quad\text{with}\quad A_X(\xi):=\sum_{x\in X}e^{2\pi i\xi\cdot x}\ne0.

The diffraction weights are nonnegative multiples of ∣AX(ξ)∣2|A_X(\xi)|^2. Thus

K(P)=min⁡{∥ξ∥:ξ∈L∗∖{0}, AX(ξ)≠0}.(2)K(P)=\min\{\lVert\xi\rVert:\xi\in L^*\setminus\{0\},\ A_X(\xi)\ne0\}. \tag*{(2)}

For a cubic period L=2ZnL = 2\mathbb{Z}^{n}, put

aX(u)=∑x∈Xeπiu⋅x,u∈Zn.a_X(u) = \sum_{x\in X} e^{\pi i u\cdot x}, \qquad u \in\mathbb{Z}^{n}.

Then K(P)=12min⁡{∥u∥:u≠0, aX(u)≠0}K(P) = \frac{1}{2}\min\{\lVert u\rVert: u \ne0,\ a_X(u) \ne0\}, and ρ(P)=N/2n\rho(P) = N/2^n. When X⊂{0,1}nX \subset\{0,1\}^{n}, the amplitude depends only on uu mod 22.

More generally, for B∈GLn(R)B \in\mathrm{GL}_n(\mathbb{R}) and PB=B(X+2Zn)P_B = B(X + 2\mathbb{Z}^{n}) the density is N/(2n∣det⁡B∣)N/(2^n|\det B|). Set

QB=∣det⁡B∣2/nB−1B−T,μB=min⁡u∈Zn∖{0}aX(u)≠0uTQBu.Q_B = |\det B|^{2/n}B^{-1}B^{-\mathsf{T}}, \qquad\mu_B = \min_{\substack{u\in\mathbb{Z}^{n}\setminus\{0\}\\a_X(u)\ne0}} u^{\mathsf{T}}Q_Bu.

The matrix QBQ_B has determinant one, and direct substitution into (1) gives

Δst(PB)=1N(μB4)n/2.(3)\Delta_{\mathrm{st}}(P_B) = \frac{1}{N}\left(\frac{\mu_B}{4}\right)^{n/2}. \tag*{(3)}

These statements do not require LL to be the primitive period lattice.

Orthogonal arrays and the Hadamard construction

For a binary motif C⊂F2nC \subset\mathbb{F}_2^n, let

1C^(u)=∑x∈C(−1)u⋅x,u∈F2n.\widehat{1_C}(u) = \sum_{x\in C}(-1)^{u\cdot x}, \qquad u \in\mathbb{F}_2^n.

Let d⊥(C)d^{\perp}(C) denote the minimum Hamming weight of a nonzero character uu for which 1C^(u)≠0\widehat{1_C}(u) \ne0, with the convention d⊥=+∞d^{\perp} = +\infty if none exists. A binary orthogonal array of strength tt is a set with all tt-coordinate projections uniform; equivalently, the character sums of weights 1,…,t1,\ldots,t vanish.

Proposition 3.1 (Exact binary Fourier gap). For a nonempty distinct binary motif CC,

K(C+2Zn)=12min⁡{4,d⊥(C)},Δst(C+2Zn)=(min⁡{4,d⊥(C)}/4)n/2∣C∣.K(C+2\mathbb{Z}^{n}) = \frac{1}{2}\sqrt{\min\{4,d^{\perp}(C)\}}, \qquad\Delta_{\mathrm{st}}(C+2\mathbb{Z}^{n}) = \frac{\left(\min\{4,d^{\perp}(C)\}/4\right)^{n/2}}{|C|}.

Proof. The shortest integer vector in a nonzero parity class has squared length equal to its Hamming weight. A nonzero vector in the zero parity class has squared length at least four; the vectors ±2ei\pm2e_i realize this and have amplitude ∣C∣|C|. Formula (2) and (1) conclude the proof. □

Let HH be a Hadamard matrix of order 12, so HH has ±1\pm1 entries and HHT=12IHH^{\mathsf{T}} = 12I. Convert the rows of the 24×1224 \times12 matrix (H−H)\begin{pmatrix} H \\ -H \end{pmatrix} to binary by mapping a sign ss to (1−s)/2(1-s)/2. Let CnC_n denote any projection onto nn columns, n=10,11,12n = 10,11,12.

Proof of Theorem 1.1. Antipodal pairing kills the character sum for every odd number of distinct columns. Hadamard column orthogonality kills the weight-two sums. Thus CnC_n is a strength-three orthogonal array. Distinct nonantipodal sign rows of HH are at Hamming distance 6 (the antipodal pairs have distance 12); deletion of at most two columns cannot identify them. Therefore ∣Cn∣=24|C_n| = 24 and Proposition 3.1 yields K=1K = 1 and Δst=1/24\Delta_{\mathrm{st}} = 1/24.

If PnP_n were a translate of a lattice, translate one point to the origin. The resulting lattice would contain 2Zn2\mathbb{Z}^n and lie in Zn\mathbb{Z}^n, so its quotient by 2Zn2\mathbb{Z}^n would be a subgroup of F2n\mathbb{F}_2^n of order 24. Every subgroup of F2n\mathbb{F}_2^n has power-of-two order, a contradiction. The numerical packing comparisons are recorded in § 7.

An explicit integral certificate

Let χ\chi be the quadratic character of F11\mathbb{F}_{11}, with χ(0)=0\chi(0) = 0. Index a 12×1212 \times12 matrix SS by {∞}∪F11\{\infty\} \cup\mathbb{F}_{11} and define S∞,∞=0S_{\infty,\infty} = 0, S∞,j=1S_{\infty,j} = 1, Sj,∞=−1S_{j,\infty} = -1, and Si,j=χ(j−i)S_{i,j} = \chi(j-i). Then ST=−SS^{\mathsf{T}} = -S, SST=11ISS^{\mathsf{T}} = 11I, and H=I+SH = I + S satisfies HHT=12IHH^{\mathsf{T}} = 12I. This is the classical Paley construction [4]. The accompanying script verifies these identities, all projections, and all character cancellations by exact integer arithmetic.

The sharp cardinality theorem

Proof of Theorem 1.2. Fix j∈{1,…,n}j \in\{1,\ldots,n\} and take the 2n2n integer-frequency indices

Bj={0,ej}∪{ei,ei+ej:i≠j}.B_j = \{0,e_j\} \cup\{e_i,e_i+e_j:i \ne j\}.

Every nonzero difference of two such indices has squared Euclidean norm at most three. Since K(P)≥1K(P) \ge1, the evaluation columns

Vb=(eπib⋅x)x∈X∈CN,b∈Bj,V_b = \left(e^{\pi i b \cdot x}\right)_{x \in X} \in\mathbb{C}^{N}, \qquad b \in B_j,

are mutually orthogonal and have norm N\sqrt{N}. Hence N≥2nN \ge2n.

If N=2nN = 2n, these columns form a scaled unitary matrix, and the rows are orthogonal. Take distinct x,y∈Xx,y \in X and set wi=eπi(xi−yi)w_i = e^{\pi i(x_i-y_i)}, so ∣wi∣=1|w_i| = 1. Factoring the row inner product using the pivot jj gives

(1+wj)(1+∑i≠jwi)=0(1≤j≤n).(4)(1+w_j)\left(1+\sum_{i \ne j}w_i\right)=0 \qquad(1 \le j \le n). \tag*{(4)}

Put W=1+∑iwiW = 1 + \sum_i w_i. Each wjw_j is either −1-1 or WW. If all are −1-1, then x−yx-y is the all-ones binary vector modulo 2. Otherwise let r≥1r \ge1 be the number of indices with wi≠−1w_i \ne-1; their common value a=Wa = W has modulus one. Counting in WW gives

(r−1)a=n−r−1.(r-1)a = n-r-1.

For r=1r = 1 this forces n=2n = 2, excluded by assumption. Hence aa is real and a=±1a = \pm1; because a≠−1a \ne-1, it equals 1, and therefore r=n/2r = n/2. Thus every difference x−yx-y is binary modulo 2 and has Hamming weight n/2n/2 or nn.

Translate a selected representative to zero; all representatives then lie in {0,1}n\{0,1\}^n modulo 2. Their 2n2n sign vectors have pairwise inner products either 0 or −n-n. A sign line contains at most two distinct vectors and there are at most nn mutually orthogonal lines in Rn\mathbb{R}^n. Therefore there are exactly nn lines, each represented with both signs. Choosing a representative from each line yields an order-nn Hadamard matrix. The converse follows from Proposition 3.1. □

Lemma 4.1 (A stronger rank obstruction). If P=X+2ZnP=X+2\mathbb{Z}^n has K(P)>1K(P)>1, then N≥Dn:=1+n+(n2)N\ge D_n:=1+n+\binom{n}{2}.

Proof. Use the frequencies 0,ei,ei+ej0,e_i,e_i+e_j for 1≤i<j≤n1\le i<j\le n. Their nonzero pairwise differences have squared norms at most four. Under the strict inequality K>1K>1 all resulting evaluation columns are mutually orthogonal. There are DnD_n such columns. □

Two exact optimality consequences

Theorem 5.1 (Bounded-complexity cubic optimality). Let n≥5n\ge5 and let P=X+2ZnP=X+2\mathbb{Z}^n have N<DnN<D_n distinct representatives at arbitrary real positions. Then

Δst(P)≤12n.\Delta_{\mathrm{st}}(P)\le\frac{1}{2n}.

Equality is possible only for N=2nN=2n, when Theorem 1.2 classifies PP up to common translation as an antipodal Hadamard array.

Proof. By Lemma 4.1, K(P)≤1K(P)\le1. Let s=4K(P)2∈{1,2,3,4}s=4K(P)^2\in\{1,2,3,4\}. For s=1s=1, trivially N≥1N\ge1. For s=2s=2, orthogonality of the columns for 0,e10,e_1 yields N≥2N\ge2. For s=3s=3, the n+1n+1 characters 0,e1,…,en0,e_1,\ldots,e_n give N≥n+1N\ge n+1. For s=4s=4, Theorem 1.2 gives N≥2nN\ge2n. Consequently,

Δst(P)≤max⁡{2−n,2−n/22,(3/4)n/2n+1,12n}.(5)\Delta_{\mathrm{st}}(P)\le\max\left\{2^{-n},\frac{2^{-n/2}}{2},\frac{(3/4)^{n/2}}{n+1},\frac{1}{2n}\right\}. \tag*{(5)}

For n≥5n\ge5 the first three terms are strictly smaller than 1/(2n)1/(2n) (the third follows by checking n=5n=5 and observing that 2n(3/4)n/2/(n+1)2n(3/4)^{n/2}/(n+1) decreases). Equality therefore requires s=4s=4 and N=2nN=2n, as asserted. □

Corollary 5.2 (The dimension-12 cubic-cell threshold). Among all equally weighted periodic configurations with marked period 2Z122\mathbb{Z}^{12} and fewer than D12=79D_{12}=79 representatives in that cell, the largest possible stealth center density is 1/241/24. Equality holds precisely for translated 24-point antipodal Hadamard configurations.

Theorem 5.3 (All binary motifs in dimensions 10–12). For n∈{10,11,12}n\in\{10,11,12\},

max⁡∅≠C⊆F2nΔst(C+2Zn)=124.\max_{\varnothing\ne C\subseteq\mathbb{F}_2^n}\Delta_{\mathrm{st}}(C+2\mathbb{Z}^n)=\frac{1}{24}.

Equality holds exactly for the binary orthogonal arrays of strength three with 24 distinct rows.

Proof. Proposition 3.1 reduces the minimum surviving squared norm to s≤4s\le4. In the cases s=1,2,3s=1,2,3 the first three bounds in (5) are strictly below 1/241/24 for n=10,11,12n=10,11,12. If s=4s=4, all binary character sums of weights one, two, and three vanish, so CC is a strength-three orthogonal array. Theorem 1.2 gives ∣C∣≥2n|C| \ge2n, and uniformity on any three coordinates forces 8∣∣C∣8 \mid|C|. Thus ∣C∣≥24|C| \ge24. Proposition 3.1 yields the upper bound and equality characterization, and our projected Paley arrays attain it. □\square

Local optimality of the marked configuration

Put

V={2ei:1≤i≤n}∪{∑i∈Iεiei:∣I∣=4, εi∈{−1,1}}.\mathcal{V}=\left\{2e_i:1\le i\le n\right\}\cup\left\{\sum_{i\in I}\varepsilon_i e_i:|I|=4,\ \varepsilon_i\in\{-1,1\}\right\}.

Lemma 6.1 (Strict reciprocal-Gram maximum). For n≥4n\ge4, the function m(Q)=min⁡v∈VvTQvm(Q)=\min_{v\in\mathcal{V}}v^{\mathsf{T}}Qv on positive-definite determinant-one symmetric n×nn\times n matrices has a strict local maximum at Q=IQ=I, with m(I)=4m(I)=4.

Proof. Let EE be any nonzero trace-free real symmetric matrix. If some Eii<0E_{ii}<0, then (2ei)TE(2ei)<0(2e_i)^{\mathsf{T}}E(2e_i)<0. If no diagonal entry is negative, all diagonal entries vanish. Choose a four-element set II containing a nonzero off-diagonal entry. Over the 16 choices of signs, the quadratic forms (∑i∈Iεiei)TE(∑i∈Iεiei)\left(\sum_{i\in I}\varepsilon_i e_i\right)^{\mathsf{T}}E\left(\sum_{i\in I}\varepsilon_i e_i\right) have mean zero and are not all zero because the degree-two sign characters are linearly independent. Hence at least one is strictly negative. By compactness on the unit sphere of trace-free symmetric matrices, there is cn>0c_n>0 with

min⁡v∈VvTEv≤−cn∥E∥F.\min_{v\in\mathcal{V}}v^{\mathsf{T}}Ev\le-c_n\lVert E\rVert_F.

Write Q=I+FQ=I+F with det⁡Q=1\det Q=1 and put E=F−(tr⁡F/n)IE=F-(\operatorname{tr}F/n)I. The expansion log⁡det⁡(I+F)=0\log\det(I+F)=0 yields tr⁡F=O(∥F∥F2)\operatorname{tr}F=O(\lVert F\rVert_F^2), and all vectors in V\mathcal{V} have squared norm four. Thus

m(Q)≤4−cn∥E∥F+4ntr⁡F<4m(Q)\le4-c_n\lVert E\rVert_F+\frac{4}{n}\operatorname{tr}F<4

for sufficiently small nonzero FF. This is a strict local maximum. □\square

Proof of Theorem 1.3. On any set of four distinct coordinates, a strength-three array has vanishing binary Fourier coefficients of weights one through three. If its weight-four coefficient also vanished, Fourier inversion would make that four-coordinate marginal uniform, hence 16∣N16\mid N, contrary to N≡8(mod16)N\equiv8\pmod{16}. Therefore every four-coordinate coefficient is nonzero. Every 2ei2e_i has amplitude NN, and changing the signs of the four nonzero coordinates leaves the relevant binary residue class unchanged. Thus aC(v)≠0a_C(v)\ne0 for all v∈Vv\in\mathcal{V}.

This finite set of amplitudes remains nonzero under all sufficiently small perturbations of the motif. The formula (3) then implies

Δst(B(C′+2Zn))≤1N(m(QB)4)n/2.\Delta_{\mathrm{st}}(B(C'+2\mathbb{Z}^n))\le\frac{1}{N}\left(\frac{m(Q_B)}{4}\right)^{n/2}.

Lemma 6.1 makes the right-hand side strictly less than 1/N1/N for a small nonsimilarity deformation of BB, while at B=IB=I, C′=CC'=C the value is 1/N1/N. The condition QB=IQ_B=I is equivalent to BB being a similarity.

For n=12n = 12 and N=24N = 24, after removing this similarity, equality forces the radius to equal one. By Theorem 1.2, the motif is a common translate of an antipodal binary Hadamard array. There are only finitely many labelled binary arrays of the specified size; in a sufficiently small neighborhood only the original remains after a chosen representative is aligned. Thus equality is locally isolated modulo the stated symmetries. □

Remark 6.2. The proof does not cover arbitrary supercell perturbations or global optimization of period shape. In dimensions 10 and 11 the theorem asserts local maximality but not isolated equality in motif coordinates.

Formal duality and numerical comparisons

Proposition 7.1 (Absence of an ordinary periodic formal dual). None of the configurations PnP_n, n=10,11,12n = 10, 11, 12, has an ordinary periodic formal dual among simple equally weighted periodic configurations in the sense of [3].

Proof. Suppose QQ were formally dual to PnP_n. Formal duality converts the normalized pair correlation of QQ into the normalized Fourier transform of that of PnP_n. The latter is supported on 12Zn\frac{1}{2}\mathbb{Z}^n, so the former is supported on Zn\mathbb{Z}^n. Since each difference of two points of a simple periodic configuration contributes positively to its pair measure, after translation Q⊆ZnQ \subseteq\mathbb{Z}^n. For each u∈Znu \in\mathbb{Z}^n, the Fourier-side pair coefficient equals one, since aCn(2u)=24a_{C_n}(2u) = 24. On the QQ side this coefficient is the fraction of points translated to another point by uu; equality to one forces uu to be a period. Hence QQ is a union of an integer number of cosets of Zn\mathbb{Z}^n and its particle density is an integer. Formal duality requires ρ(Q)ρ(Pn)=1\rho(Q)\rho(P_n) = 1, hence ρ(Q)=2n/24\rho(Q) = 2^n/24, which is not an integer in the three stated dimensions. Contradiction. □

The following recorded lattice center densities are those of the duals of the best known lattices in the respective dimensions [2]:

nnLattice center densityΔst(Pn)\Delta_{\mathrm{st}}(P_n)RatioUnit-density KK
101/(163)1/(16\sqrt{3})1/241/242/32/\sqrt{3}1.455488812…1.455488812\ldots
111/(2⋅35/2)1/(2 \cdot3^{5/2})1/241/2435/2/123^{5/2}/121.498153270…1.498153270\ldots
121/271/271/241/249/89/81.534660772…1.534660772\ldots

Table 1.

Cohn’s table lists overall known packing center densities 5/1285/128, 9/2569/256, and 1/271/27 in dimensions 10, 11, and 12, respectively. These are comparisons of the dimensionless normalized quantities, not constructions of better sphere packings. The optimal packing density among all lattices in these dimensions is not known, so this paper establishes no separation from every lattice.

Reproducibility, provenance, and scope

The accompanying standard-library Python script constructs the Paley order-12 matrix, verifies HHT=12IHH^{\mathsf{T}} = 12I exactly, checks the 24 distinct projected rows and the weight-1,2,3 character cancellations, and enumerates weight-four values. It is a finite-instance verification of the construction, not a machine proof of Theorems 1.2–1.3. The statements of the rank and local results are given with full mathematical proofs above. The verification script verify_hadamard.py and its output verification.txt are available among the arXiv submission’s ancillary files.

The underlying Hadamard arrays are classical and should not be claimed as a new construction. Establishing publication priority for the stealth application, the arbitrary-position equality classification, and local marked-period rigidity requires specialist comparison with existing cubature, design, and hyperuniformity literature. This manuscript is an AI-assisted draft (GPT 6-Astra) requiring final author approval.

In dimension two, a recent AI-generated manuscript released by OpenAI [7] claims that the triangular lattice is universally optimal among planar configurations of density one. That result concerns energy minimization rather than stealth, and we do not use it here. Whether it yields a corresponding statement that a lattice is stealth-optimal in the plane, and whether such claims can be verified with AI-assisted formal proof, remain open.

Acknowledgments

The author thanks xAI for providing a SuperGrok subscription at no cost, and the Turing Fellowship for providing a Claude Max subscription for free.

References

  1. [1]A. Cohen, H. Cohn, and M. Viazovska, The stealthiest particle configurations in 8 and 24 dimensions, arXiv:2609.03121 (2026). https://arxiv.org/abs/2609.03121.
  2. [2]H. Cohn, Sphere packing: tables of best known densities and bounds, https://cohn.mit.edu/sphere-packing/, accessed November 2026.
  3. [3]H. Cohn, A. Kumar, C. Reier, and A. Schürmann, Formal duality and generalizations of the Poisson summation formula, Contemporary Mathematics 625 (2014), 123–140. https://arxiv.org/abs/1306.6796.
  4. [4]R. E. A. C. Paley, On orthogonal matrices, Journal of Mathematics and Physics 12 (1933), 311–320.DOI
  5. [7]OpenAI, Universal optimality of the triangular lattice, manuscript dated 23 September 2026, released 6 October 2026. https://github.com/openai/math.

Paper details

vlnc:W_2vgnu6uy
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY-SA 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Mathematics

Not recorded for this paper

View statement
Looking for related papers…
Contents