arXiv++ Combinatorics

The 2026 disproof of Erdős's unit-distance conjecture and Sawin's subsequent explicit quantitative refinement show that the maximum number $u(n)$ of unit distances among $n$ planar points can exceed $n^{1+\varepsilon}$ for a fixed positive $\varepsilon$. Sawin's explicit bound gives more than $n^{1.014}$ unit distances for arbitrarily large $n$ and exposes integer parameters whose choice is not fully optimized. This report starts from Sawin's nonlinear integer optimization problem and develops an open-source Python verification pipeline. The pipeline is first validated by reproducing Sawin's published parameter choice and is then applied to computationally improved certificates. We optimize and verify certificates involving sets of primes $T$ and $S_Q$, integer multiplicities $k(p)$, and a rationally encoded real parameter $R$. The implementation is deliberately lean, so that all results can be replicated on standard hardware and the procedures can be extended. We compare a deterministic greedy heuristic, a tailored integer evolution strategy with two-sided geometric, or discrete-Laplace, integer mutation and repair operators for number-theoretic feasibility, and a two-parent discrete-recombination variant. Four certificate levels are reported: Sawin's published example with $δ=0.0141144286784982\ldots$, a greedy certificate with $δ=0.0151718056372133\ldots$, a tailored integer evolution strategy certificate with $R=6672416/100000$ and $δ=0.0152616610684193\ldots$, and a recombination variant with the same $R$ and $δ=0.0152628688170072\ldots$. Consequently, the best current certificate supports the cautious statement $u(n)>n^{1.0152}$ for arbitrarily large $n$. Beyond this unit-distance application, the work illustrates how randomized optimization heuristics can improve explicit certificates in pure mathematics and combinatorial geometry.

A long-standing question is whether chromatic symmetric functions can distinguish non-isomorphic trees. Campbell introduced chromatic noncommutative symmetric functions for digraphs, which lift chromatic symmetric functions to NSym, and asked to what extent they can distinguish non-isomorphic oriented trees. In this article, we prove that chromatic noncommutative symmetric functions can reconstruct oriented stars, oriented double stars, some oriented caterpillars, and some oriented paths.

Regev and Stephens-Davidowitz conjectured that the integer lattice maximizes Gaussian mass among integral lattices of a given rank. We prove this, including the equality case, for all unimodular Construction-A lattices arising from binary self-dual codes. The proof reduces the theta-series inequality to a sharp majorization statement for codes: if $C$ is a binary self-dual $[2k,k]$ code, then the half-weight distribution of $C$ is dominated in convex order by $\operatorname{Bin}(k,1/2)$, which is the corresponding distribution for the repetition-code model of $\mathbb{Z}^{2k}$. Indeed, after putting $C$ in systematic form $[I\mid A]$, self-duality gives $AA^T=I$ over $\mathbb{F}_2$, so for a uniformly random message $a$ the two weights $\operatorname{wt}(a)$ and $\operatorname{wt}(aA)$ have the same binomial law. The half-weight of the resulting codeword is their average, and Jensen's inequality then gives convex-order domination. Applied to the convex test functions that build the theta series, this yields a sum-of-squares formula for the Gaussian-mass gap; applied to hinge functions, it gives coefficientwise nonnegativity of the reduced gap polynomial.

It is well-known that any class of simple graphs, that is characterized by finitely many forbidden minors, also admits a characterization by finitely many forbidden topological minors; furthermore, the list of forbidden topological minors may be derived from the list of forbidden minors. We prove a similar result in Matching Theory. Our Main Theorem states that any class of matching covered graphs, that is characterized by finitely many forbidden $S$-minors that are cubic, also admits a characterization by finitely many forbidden conformal minors that are cubic as well; once again, the list of forbidden conformal minors may be derived from the list of forbidden $S$-minors. In order to establish the above, we first prove that every matching covered graph has one of two graphs as a conformal minor -- either $K_4$, or the $Θ$ graph (that is, two vertices joined by three edges). (In fact, we need and prove a much stronger statement.) This is reminiscent of a theorem due to Lovász: every nonbipartite matching covered graph has one of two graphs as a conformal minor -- either $K_4$, or the triangular prism $\overline{C_6}$. As applications of our Main Theorem, we deduce known 'forbidden conformal minor characterizations' of pfaffian near-bipartite graphs, and of pfaffian solid graphs, using their respective known 'forbidden $S$-minor characterizations'.

Graph burning is a deterministic discrete-time process that models the propagation of information within a network as a set of fires that spread in a graph. The associated graph parameter, the burning number, measures the speed of that spread. The smaller the burning number of a graph is, the faster information spreads in the corresponding network. In this paper, we mainly study the burning number from a structural point of view. In particular, we structurally characterize all graphs with burning number 2. We then proceed to compute the burning number of split graphs and $\{ K_2+2K_1, P_4, C_4\}$-free graphs. In particular, we show that connected split graphs and connected $\{ K_2+2K_1, P_4, C_4\}$-free graphs are well-burnable. Lastly, we provide a blueprint on how to investigate the burning number of a given social network.

Motivated by Norine's conjecture, this paper investigates monochromatic antipodal paths in $2$-edge-coloured hypercubes. Our main method relies on a topological criterion applied to triangulated $2$-skeleta. We show that any antipodal colouring of a centrally symmetric, simply connected $2$-complex yields a monochromatic path linking an antipodal pair of vertices. By symmetrically triangulating opposite square faces in certain classes of colourings, we obtain a topological verification of Norine's conjecture for these classes. We also establish quantitative bounds for cases in which only a limited number of square faces present structural obstructions.

This paper considers holomorphic modular forms for $Γ_1(N)$ of integral weight of the form $$f^{(N)}_{\mathbf a}(τ) =q^{s} (q^{N};q^{N})_{\infty}^{a_0}\prod_{j=1}^{\lfloor N/2 \rfloor}(q^j,q^{N-j};q^N)_\infty^{a_j}, \quad \mathbf a = (a_1, \ldots, a_{\lfloor N/2 \rfloor}),$$ for fixed $a_0=2k \in 2 \Bbb Z_{\ge 0}$. We show that the number of relevant exponent vectors $\mathbf a$ is finite and characterize them in terms of the $\mathbb{Q}$-rational cuspidal divisor class group of $X_{1}(N)$. Effective procedures are given for counting the admissible exponents by enumerating the corresponding polytopes. This leads to formulas for the number of exponent vectors in terms of quasipolynomials in $k$.

While the uncrowding algorithm on set-valued tableaux has long been instrumental in proving the Schur positivity of stable symmetric Grothendieck polynomials, lattice models have emerged as a modern framework for investigating symmetric functions, in particular symmetric Grothendieck polynomials. In this work, we synthesize these combinatorial and lattice-theoretic approaches by defining both the Robinson--Schensted--Knuth (RSK) correspondence and the uncrowding operation directly on a 5-vertex model of Motegi and Sakai and its subsequent reinterpretation by Buciumas, Scrimshaw, and Weber. Our lattice-based RSK formulation yields a powerful new result: the direct construction of the associated crystal structure on the states of the 5-vertex model.

We prove a conjecture of Glasby, Praeger, and Unger concerning the symmetric group $S_{n}$. Let $π_{n}$ denote the proportion of elements of $S_{n}$ that are pre-$p$-cycles for some prime $p\in[2, n-3]$. We prove that $π_{n} > 1/3$ for all $n\geq 8$.

The partial Petrial polynomial of a ribbon graph, introduced by Gross, Mansour and Tucker, enumerates partial Petrials by Euler genus. Recently, Deng, Jin and Yan defined an analogue for grafts and showed that it can be expressed as a rank-generating function of an adjacency matrix. In this paper we first prove a recurrence relation that reduces the partial Petrial polynomial of a graph with respect to an arbitrary edge, expressing it as a sum of three terms involving graphs obtained by local complementation and edge pivoting. This recurrence extends the known leaf-reduction formula to vertices of any positive degree. Second, using this recurrence we compare the lowest and highest degree coefficients of the polynomial. We prove that the lowest coefficient is always at most the highest coefficient, and that equality holds if and only if the graph has no edges.

The Membership Problem for Pedigree Polytope (M3P) asks, given $X\in\mathbb{Q}^{\binom{n}{3}}$, whether $X\in\mathrm{conv}(P_n)$, where $P_n$ is the set of all pedigrees. A pedigree is a structured encoding of a Hamiltonian cycle construction in $K_n$. We establish that M3P is solvable in strongly polynomial time via a recursively constructed layered network $(N_k, R_k, μ)$ and a multicommodity flow problem MCF$(k)$. The necessary and sufficient condition for membership established is that the optimal total flow in MCF$(n-1)$ equals the maximum possible flow $z_{\max}$. The complexity analysis, grounded in Tardos's strongly polynomial algorithm for combinatorial linear programs (1986), shows that this condition can be checked in strongly polynomial time in the dimension of the matrix involved. By sufficiency, this implies M3P~$\in$~P. Since the Symmetric Travelling Salesman Problem (STSP) reduces to M3P via the Multistage Insertion (MI) formulation (Arthanari 1983), STSP is solvable in polynomial time, and the P vs.NP question is resolved. The proofs leading to this result are fully machine-verified in Lean~4/Mathlib4, with zero unresolved \texttt{sorry}s in the main proof chain. The main contribution is the Lean~4 machine verification of all proofs in the main chain, resulting in \texttt{theorem p\_equals\_np}: P = NP. The Lean~4 formal verification covers the sufficiency of MCF(n-1) for membership in $\mathrm{conv}(P_n)$, and the P = NP chain via Maurras (2002), Grötschel--Lovász--Schrijver (1988), Cook (1971), and Karp (1972). The complete lean project (36 Lean~4 files, 2968/2968 build targets clean) is available at https://github.com/TiruArt/Pedigree-Polytopes-Lean4.

Efforts have been made to extend Cassini's identity (also known as Simson's identity) to the k-step or k-bonacci numbers for decades. These efforts have lacked both completeness of result and simplicity of proof, and this question remains open and relevant. In this note, we offer a definitive solution as well as the generalization of both Catalan's and Vajda's identities.

In this paper, we consider triangle-free graphs with diameter 2. If a triangle-free graph $G$ with diameter 2 is not isomorphic to a star, then the radius of $G$ is also 2, where such a graph is also called a $2$-self-centered graph. Shekarriz et al. [A characterization for 2-self-centered graphs, Discuss. Math. Graph Theory 38 (2018), 27--37.] gave a characterization of 2-self-centered graphs. However, there is a slight flaw in their characterization. Thus, in this paper, we modify it and prove an accurate characterization of those graphs. Furthermore, by using our characterization, we prove some results concerning the chromatic number of triangle-free graphs with diameter 2.

Johnson graph are a family of graphs that play an important role in the theory of constant-weight codes, extremal combinatorics, and combinatorial geometry. We study signed analogues of classical Johnson graphs, denoted by $J_\pm(n,k)$, whose vertices are vectors of the form $\pm e_{i_1}\pm\cdots\pm e_{i_k}$, where two vertices are adjacent whenever their dot product equals $k-1$. We are particularly interested in maximum independent sets in the case $k=4$. An example of such an independent set in $J_\pm(n,4)$, which we call \emph{classical}, is obtained by lifting an arbitrary optimal $(n,4,4)$-code. Such independent sets naturally define kissing arrangements in ${\mathbb R}^n$. We develop an algorithm that is practical for computing all maximum independent sets in $J_\pm(n,4)$ up to signed permutations for $n\le 12$, $n\ne 11$. In addition to obtaining complete lists, we provide structural characterizations of all types of maximum independent sets in these dimensions, excluding $n=5$ and $n=11$. Our most striking results concern the case $n=12$. We identify $1579$ non-isomorphic maximum independent sets in $J_\pm(12,4)$, all corresponding to non-isometric kissing arrangements of size $840$ in ${\mathbb R}^{12}$. Structurally, $1575$ of these independent sets arise from three different constructions, the rest are liftings of one of four $(12,4,4)$-codes. To our knowledge, this is the first dimension in which such a large diversity of potentially optimal kissing arrangements has been observed. Beyond this finite range, we prove that for $n\equiv 2$ or $4 \pmod 6$, every maximum independent set arises from a Steiner quadruple system. We also obtain a characterization of the so-called \emph{nontrivially self-compatible} codes, namely optimal $(n,4,4)$-codes from which non-classical maximum independent sets can be constructed.

We prove an upper bound for the number of subgroups of an abelian $p$-group, with constants that are sharp on homocyclic blocks. If $A_λ$ has type $λ$, the main exponent is $η(λ)=\sum_i\lfloor(λ'_i)^2/4\rfloor$, where the parts of $λ'$ are the column heights of the Ferrers diagram of $λ$. The leading coefficient is the number of central choices in the blocks of columns of odd height. The proof starts from Delsarte's formula, separates the blocks by column height, and estimates each homocyclic block by successive Durfee squares. We also give three applications: a comparison theorem for $p$-groups of nilpotency class less than $p$, diagonal summatory estimates with the shape fixed, and formulae for leading terms at fixed rank for finite abelian $p$-groups.

We present a framework to generalize the standard b-ary Gray code to get the k-bonacci ones obtained in [5] as well as many others by using theoretical tools that allow to make calculations on lists. We introduce the notion of Z-Gray product, from which we deduce sequences of lists of finite words avoiding a predefinite list Z of factors and which satisfy a power-associativity property as well a generalizations of the classical flipping digit property.

Building on the recently discovered origami-amplituhedron correspondence, we prove that the BCFW (Britto-Cachazo-Feng-Witten) cells triangulate the $m=4$ amplituhedron in full generality at all loop orders, both in momentum and momentum-twistor space. Along the way, we develop two natural "$L$-punctured" extensions of the positive Grassmannian and relate them via T-duality.

We study finite extremal problems for palettes, which arise from the palette framework for the uniform Turán densities of $3$-uniform hypergraphs. Recent work has developed reductions from palette colorability questions to extremal problems for digraphs. In this paper we prove an exact degree-square refinement of these reductions for a natural family of left and right tournament palettes. For a tournament $T$, let $P_T^L$ and $P_T^R$ denote the left and right palettes generated by $T$. We prove that if $T$ is self-converse and has at least two vertices, then for every $m\ge 1$ the maximum number of admissible triples in an $m$-color palette avoiding both $P_T^L$ and $P_T^R$ is \[ \operatorname{ex}_2^+(m,T) = \max\left\{ \sum_{v\in V(D)} d_D^+(v)^2: |V(D)|=m,\; D\text{ is }T\text{-free} \right\}. \] The proof attaches two auxiliary digraphs to each palette and converts the palette optimization into a degree-square Turán problem. We also prove a general majorization principle for convex out-degree moments in $F$-free digraphs. Whenever an ordinary Turán extremal construction has extremal initial segments, the same construction maximizes every nondecreasing convex function of the out-degree sequence. Applying this to the Brown--Harary and Zhou--Li extremal digraphs for directed cycles gives exact formulas for all convex out-degree moments in $\overrightarrow{C}_{\ell}$-free digraphs. In particular, $\operatorname{ex}_2^+(m,\overrightarrow{C}_{3}) =\frac{m(m^2-1)}{3}.$ Consequently, for $m$ color the sharp density avoiding the two cyclic-triangle palettes is $\frac13-\frac1{3m^2}.$ Combining this exact finite endpoint with the palette classification theorem, we obtain finite $3$-graphs $H_m$ satisfying \[ \frac13-\frac1{3m^2} \le π_{\mathrm u}(H_m) \le \frac13. \] Thus the densities of these finite hypergraphs converge to $\frac13$.

The neighborhood complex $N(G)$ and the dominance complex $D(G)$ are fundamental simplicial complexes associated with a graph $G$. We characterize precisely when the Stanley-Reisner ring $k[N(G)]$ admits a $2$-linear resolution, thereby answering an open question posed by Fröberg. We prove that this occurs if and only if $G$ is neighborhood conformal and its common neighbor graph is chordal. Equivalently, $G$ is a bipartite graph whose only induced cycles are $4$-cycles. As a consequence, we show that Katzman's lower bound becomes an equality for this class, yielding $\operatorname{reg}(S/I(G))=\operatorname{im}(G)$. Using recent results on glued clique complexes, we derive explicit combinatorial formulas for the exact graded Betti numbers of these neighborhood complexes. Finally, utilizing combinatorial Alexander duality, we obtain a corresponding classification of Cohen-Macaulay dominance complexes, and prove that the dominance complex of a graph without isolated vertices admits a $2$-linear resolution if and only if the graph is a star graph.

For a possibly ramified $\mathbb{Z}_p^d$-covering of connected graphs, we establish an Iwasawa-type asymptotic formula for the growth of the $p$-adic valuations of the complexities. The formula is expressed as a polynomial in $n$ and $p^n$ with explicit leading coefficients $λ$ and $μ$; in particular, we eliminate the error term of the form $O(p^{(d-1)n})$ appearing in earlier work. We then establish a Kida-type formula describing the behavior of $λ$ under a $p$-covering between $\mathbb{Z}_p^d$-coverings, assuming $μ= 0$. Finally, for any fixed $p$ and integer $d \geq 2$, we construct an unramified $\mathbb{Z}_p^d$-covering of a bouquet with prescribed $λ$- and $μ$-invariants.

For integers $n \geq k \geq 2$ and $0 \leq a,b \leq k-1$, let $m_{k,n}(a,b)$ denote the maximum size of an $(a,b)$-town (mod $k$) family of an $n$-element set, a collection of subsets of whose cardinalities are congruent to $a$ modulo $k$ and whose pairwise intersections are congruent to $b$ modulo $k$. This notion generalizes the classical Oddtown and Eventown problems. We prove that $m_{k,n}(a,b)\leq n$ whenever $a\not\equiv b\pmod{k}$, thereby resolving a conjecture of Veselinov and Marinov. We also disprove another conjecture of theirs by showing that $m_{3,11}(2,2)>m_{3,11}(1,1)$. For the diagonal case $a\equiv b\pmod{k}$, we establish the general bound $m_{k,n}(a,a)\leq 2^{\lfloor n/2\rfloor}$ and completely determine when equality holds. We further obtain improved bounds and exact values in several special cases. The proofs combine characteristic-zero linear algebra with methods from coding theory and finite geometry.

We introduce a method based on horizontal slicing and a plane-then-line stopping-time decomposition for the prime field spherical restriction problem in four dimensions. The method is designed to overcome the Kloosterman obstruction in the spherical Bochner--Riesz kernel by decomposing each critical horizontal slice into rich-plane, rich-line-and-poor-plane, and poor-line-and-poor-plane components, which are then treated by distinct affine-geometric mechanisms. As a quantitative consequence of this structural method, we prove that \[ R_{S_j}^*(2\to r)\lesssim 1 \] for every nonzero sphere $S_j\subset\mathbb{F}^4$ and every $r>23/7$. As an application, we obtain the first improvement over the twenty-year-old $(d+1)/2$ threshold in the four-dimensional prime field Erdős-Falconer distance problem.

The $r$-neighbor bootstrap percolation process on a graph is a vertex-activation process that begins with a set of initially active vertices. In each subsequent round, every inactive vertex having at least $r$ active neighbors becomes active. A set of initially active vertices whose activation eventually spreads to all vertices of the graph is called a percolating set. Let $m(G,r)$ denote the minimum size of a percolating set in the $r$-neighbor bootstrap percolation process on a graph $G$. In this paper, among other results, we prove that \[ \frac{k^2}{4}+Ω(k)\leqslant m(\mathbbmsl{O}_k,2)\leqslant \frac{k^2}{3}+O(k), \] where $\mathbbmsl{O}_k$ denotes the odd graph on a ground set of cardinality $2k+1$. As a consequence, we confirm a conjecture posed in 2021 by Grippo, Pastine, Torres, Valencia-Pabon, and Vera.

A classical theorem of Nash-Williams states that if $G$ is a $2$-connected graph on $n$ vertices with minimum degree at least $(n+2)/3$, then for every longest cycle $C$ of $G$, the graph $G-V(C)$ is edgeless. Motivated by a higher-connectivity analogue, Bondy conjectured in 1980 that if $G$ is a $k$-connected graph on $n$ vertices with minimum degree at least $(n+k(k-1))/(k+1)$, then for every longest cycle $C$ of $G$, every path in $G-V(C)$ has at most $k-1$ vertices. This conjecture is known for $k\le 3$ and remains open for all $k\ge 4$. In this paper, we prove Bondy's conjecture for all sufficiently large graphs. The key ingredient is a new Dirac-type theorem that gives a lower bound on the length of a longest cycle in a $k$-connected graph, which also yields a partial solution to a conjecture of Jung from 1990. Along the way, we develop several new tools, including a DFS lemma and an average-degree analogue of the Bondy--Jackson theorem. We conclude with a discussion of related problems and a counterexample to a conjecture of Voss from 1991.

We use recent results, among which modular inequalities for curves, to determine the Alexander polynomials for some classes of pencil-type conic-line arrangements. For these classes of curves we prove that the Alexander polynomial is (at least partially) combinatorial. To this end, we exemplify new techniques that are suitable for broader use, lending themselves to more general classes of curves.

For an integer $k\ge 0$ and a graph $G$, the \emph{token-sliding reconfiguration graph $\mathsf{TS}_k(G)$} has the independent $k$-sets of $G$ as vertices. Two vertices are adjacent if one token can slide along an edge of $G$ and the resulting $k$-set is still independent. We study the following realizability problem: for fixed $k\ge 2$, which graphs are isomorphic to $\mathsf{TS}_k(G)$ for some graph $G$? This inverse viewpoint asks which abstract state spaces can occur exactly under a local token rule. We give positive realizability results for the complement targets $\overline{K_n}$, $\overline{K_{m,n}}$, and $\overline{K_n-e}$, and we determine sharp cutoffs for complements of paths and cycles. We also prove a product formula for token-sliding graphs of disjoint unions and apply it to Cartesian products of complete graphs, paths, and cycles. For every grid $Γ_{m,n}=P_m\square P_n$ with $2\le m\le n$, we realize $Γ_{m,n}$ at token value $m+n-2$ and at every token value $k\ge 4$. At small token values, we prove that $C_4\square C_n$ is not a $\mathsf{TS}_2$-graph for $n\ge 4$, classify ladders $Γ_{2,n}$, and settle the first non-ladder grid: for $k\ge 2$, $Γ_{3,3}$ is realizable if and only if $k\ge 4$.

We propose a novel method for computing invariant densities of certain multidimensional continued fraction algorithms. Inspired by Rauzy induction, our approach builds on the formalism of simplicial systems developed by Fougeron. We introduce a win-lose induction on a graph that is conjugate to the original algorithm, and construct its natural extension by introducing the notion of a dual graph. This method explicitly reconstructs the complete dynamics of the algorithm, yielding a partition of the invariant domain of the natural extension into pieces that map to one another. We further study the ergodic properties of the algorithms within this framework; in particular, we prove that the Modified Triangle algorithm in any dimension admits a unique ergodic measure equivalent to the Lebesgue measure.

A finite group $G$ is $\varepsilon$-quasirandom if all its nontrivial irreducible complex representations have degree at least $|G|^\varepsilon$. Building on recent work of Golsefidy-Srinivas, we prove that expansion in a quasirandom group is controlled by expansion in its simple quotients. As a consequence, we remove the product theorem from the hypotheses of the Bourgain-Gamburd expansion machine. Moreover, we combine this result with crown theory to deduce that $1 + \lfloor \varepsilon^{-1} \rfloor$ random elements give an expander Cayley graph with high probability. Finally, generalizing results of Breuillard-Green-Tao and Pyber-Szabó, we prove that the diameter of any connected Cayley graph of a quasirandom group is polylogarithmic.

In 1983, Kusner conjectured that the largest equilateral set in $\mathbb{R}^{n}$ with metric $\ell_{p}$ has cardinality $n+1$ when $1<p<\infty$ and $2n$ when $p=1.$ This conjecture was proved only in the isolated cases $p=2$ and $p=4$, and was disproved when $1<p<2$. The best general upper bound $O_p(n^{\frac{2p+2}{2p-1}})$ is due to the celebrated work of Alon and Pudlák~[GAFA, 2003]. Our main contributions include: (1) We prove Kusner's conjecture for every dimension $n\ge 1$ when $2\le p\le 4$. More generally, for every integer $k\ge 0$ and every $p\in[4k+2,4k+4]$, every equilateral set in \(\mathbb{R}^{n}\) with metric $\ell_p$ has cardinality at most $(2k+1)n+1$. On the complementary intervals $p\in(4k,4k+2)$ with $p\geq 1$, we obtain the almost linear bound $O_p(n\log n)$. (2) We also consider the analogous problem on the torus $\mathbb{T}^n$, recently initiated by Alon, where the cyclic distance makes the problem substantially more delicate than in $\mathbb R^n$. We prove the almost linear bound $O_p(n\log n)$ for $1\le p\le 2$ and $O_p(n^{\frac{3}{2}-\frac{1}{p}})$ for every fixed real $p>2$, improving Alon's bounds $O_p(n^{2+\frac{2}{\lfloor p\rfloor}})$ for all finite $p\ge 1$.