arXiv++ Combinatorics

Let $G$ be a finite simple graph. An averaging operation selects a nonempty connected vertex set and replaces the weights of its vertices by their arithmetic mean. For a nonnegative initial weight function $ω:V(G)\to\mathbb R_{\ge0}$ and a target vertex $x$, let $t(G,x,ω)$ denote the minimum number of averaging operations needed to attain the maximum reachable weight at $x$. Gollin et al.[1] proved, using the Axiom of Choice, that a finite optimal strategy exists for every finite graph, every nonnegative initial weight function and every target vertex; see Tao [2] for a simplified proof. We prove that the lengths of shortest optimal strategies on $3$-spiders have no uniform finite bound. In particular, these lengths cannot be bounded in terms of the maximum degree of the graph.

Let a finite simple graph be given. We study symmetric matrices with this graph pattern that admit a null vector with prescribed exact support. Such a vector exists precisely when its support is a fort, meaning that no vertex outside the support has exactly one neighbor inside it. Over the complex numbers, we prove that the Zariski closure of each exact-support coefficient locus is irreducible and compute its codimension in terms of boundary vertices and a labeled forest on the connected components of the support. The corresponding forest deficiency also determines the generic restricted nullity. An incidence parametrization yields a prime decomposition of the reduced zero-forcing failure locus. For a prescribed common kernel, we describe the coefficient space through equilibrium stresses, exterior affine dependencies, and free coefficients. We then construct boundary complexes recording local solution values and equation residuals. Their gluing maps are defined over the universal coefficient ring and commute with arbitrary base change. By retaining the complete collection of support signatures associated with each seed set, we obtain a composition formula for the minimum zero-forcing number and an algorithm for its computation.

We prove the 3-Decomposition Conjecture: every finite connected cubic loopless multigraph decomposes into a spanning tree, a 2-regular subgraph, and a matching. The proof rests on a new theorem on matching complements in subcubic graphs. Let H be a finite connected bridgeless simple graph of maximum degree three, and let S be its set of degree-two vertices, with |S| = k >= 2. We show that H has a matching of size equal to its cyclomatic number, |E(H)| - |V(H)| + 1, whose deletion leaves a single tree containing all of S, together with cycles disjoint from S. The proof is by induction on k.

Let $F$ be a finite simple graph with $m$ vertices and at least two edges, and let $p\in(0,1]$. We prove that a graph sequence is $p$-quasirandom if and only if the symmetrized counts of copies of $F$ across every balanced partition into $m$ classes have the asymptotics predicted by the random graph with edge probability $p$. This settles a conjecture of Janson and Sós for all such $F$, including graphs with isolated vertices. The proof completes their graph-limit reduction by a concavity argument for a sum of fractional powers determined by the degree sequence of $F$. We also show that a symmetrized subgraph kernel of positive mean can be additive in its coordinates only when the underlying graphon is constant almost everywhere. This rigidity result yields a quasirandomness criterion for generalized restricted subgraph counts, resolving a further conjecture of Janson.

We prove that the average order of connected induced subgraphs of a connected simple graph of order $n$ and minimum degree at least $3$ is strictly greater than $n/2$, which settles a conjecture of Vince~[\emph{J. Graph Theory} 97 (2021), 82--103]. We go further and show that, for connected simple graphs of order $n$ and minimum degree exactly $k\ge3$, $K_k\vee(K_1\cup K_{n-k-1})$ uniquely minimizes this average under additional conditions.

We prove that deciding whether the weak partition connectivity (WPC) of an unweighted hedgegraph is at most a given integer is NP-complete, resolving an open question about its computational complexity. Hardness holds even for connected full-support split systems, where each hedge consists of two nonempty, vertex-disjoint hyperedges covering all vertices. We also obtain hardness on connected edge-colored simple graphs in which each color class is a spanning forest with exactly two nontrivial trees. On both classes, exact computation of fractional WPC is NP-hard, whereas hedge connectivity is computable exactly in polynomial time. A binary matrix formulation yields the reduction and a deterministic polynomial-time exact algorithm under a linear intersection condition on row supports after normalization and column compression. In particular, both integer and fractional WPC are computable in polynomial time when these supports have size at most two, whereas NP-hardness holds already when their size is bounded by three. For every fixed $b$, we give deterministic and randomized efficient polynomial-time approximation schemes on hedgegraphs in which each hedge has at most $b$ connected components, counting isolated vertices. Both schemes return one partition that simultaneously approximates integer and fractional WPC. Unless $\mathrm P=\mathrm{NP}$, neither objective admits a deterministic FPTAS, even on the two hard instance classes.

An additive coloring assigns positive integer labels to the vertices of a graph so that adjacent vertices have different sums of their neighbors' labels. The additive chromatic number $η(G)$ is the least $k$ for which labels from $[k]$ suffice. We give an elementary orientation argument proving that every finite simple graph $G$ satisfies $η(G) \leq Δ(G) + 1$. This argument yields a $(Δ(G) + 1)$-additive-coloring in polynomial time and improves upon the current best bound of $η(G) \leq Δ(G)^{2} - Δ(G) + 1$, valid for $Δ(G) \geq 2$, from Akbari, Ghanbari, Manaviyat, and Zare.

Let $G=(V(G),E(G))$ be a finite connected simple graph. The edge-connectivity $λ(G)$ of $G$ is the minimum number of edges that must be removed to make $G$ disconnected. We call $G$ maximally edge-connected when $λ(G)=δ(G)$, where $δ(G)$ is the minimum degree of $G$. The square graph $G^2$ of $G$ is the graph with vertex set $V(G)$, where two distinct vertices $u$ and $v$ are adjacent in $G^2$ precisely when their distance in $G$ is at most two. In this paper, first we prove that if $λ(G)\geq4$ and $λ(G^2)<δ(G^2)$, then $λ(G^2)>λ(G)^{3/2}-\frac54λ(G)$, which extends the result in \cite{ChenXu} from the edge-connectivity range $4\leqλ(G)\leq7$ to $λ(G)\geq4$. Then we show that $G^2$ is maximally edge-connected if one of the following conditions hold: (i) $δ(G)\geq\lfloor(n+2)/4\rfloor$, or $δ(G)\leq\lfloor(n+2)/4\rfloor-1$ and $\sum_{i=1}^{2k}d_{n+1-i}\geq k(n-4)-2δ(G)+3$, where $k$ is an integer satisfying $2\leq k\leqδ(G)+2$ and $d_1\geq d_2\geq\cdots\geq d_n$ are the degrees of $G$; (ii) $d_{n-δ(G)}+d_n\geq(3n-9)/4$.

The reduced power graph $Γ_{RP}(G)$ of a group $G$ is a simple graph with vertex set $G$, where two distinct vertices $u$ and $v$ are adjacent if and only if $\langle u \rangle \subset \langle v \rangle$ or $\langle v \rangle \subset \langle u \rangle$. An $L(2,1)$-labeling of a graph $Γ$ is a mapping from the vertex set of $Γ$ to the set of non-negative integers such that the labels of adjacent vertices differ by at least $2$ and the labels of vertices at distance two are distinct. The lambda number of $Γ$, denoted by $λ(Γ)$, is defined as the minimum possible span of labels among all $L(2,1)$-labelings of $Γ$. In this paper, we establish upper and lower bounds for $λ(Γ_{RP}(G))$ and characterize the conditions under which these bounds are achieved. As an application, we determine the exact value of $λ(Γ_{RP}(G))$ for certain classes of finite groups.

We study bootstrap percolation with independent vertex thresholds taking values one and two, with threshold-one probability $(1-c/d)/d$ for fixed $c>0$. The seed set is chosen uniformly among sets of a prescribed deterministic size, independently of the graph and thresholds. We prove threshold statements at fixed relative margins. For uniform simple graphs with prescribed nonnegative integer degrees of even sum and exact mean $d$, assume $\max_i|d_i-d|\le C\sqrt d$ and $c+1-v_n\geκ>0$, where $v_n=\operatorname{Var}(D)/d$ and $C,κ$ are fixed. When $d\to\infty$ and $d=o(n^{1/7})$, the leading seed scale is $n(c+1-v_n)^2/(2d^4)$, without requiring a limit of $v_n$. At fixed relative margins below and above this scale, the final active set has size $O(n/d^3)$ and $n-o(n)$, respectively, with high probability. For regular graphs, the supercritical conclusion strengthens to full activation. For $G(n,d/n)$, the same $O(n/d^3)$ and $n-o(n)$ conclusions hold when $d\to\infty$ and $d^5/n\to0$, with seed scale $nc^2/(2d^4)$. Thus regular and independent-edge graphs have different coefficients at the same asymptotic mean degree. Local exploration estimates yield explicit inactive remainders and survive conditioning on simplicity in the prescribed model. We also quantify the precision obstruction to static inclusion transfer and compute deterministic response-barrier corrections, without identifying a shrinking random critical window.

We study the loopy polynomial L_G, a multivariate graph invariant arising from bizonotopal graphical algebras and defined by deletion and loopy contraction, in which the contracted edge is retained as a loop. It determines the Tutte polynomial and admits a spanning-forest expansion recording external activity componentwise. It also determines Stanley's chromatic symmetric function for every graph; the degree sequence, induced edge count profile, and independence polynomial for loopless graphs; and the clique and matching polynomials for simple graphs. Separating component size from external activity gives a refinement that is equivalent to the extended U-polynomial and the extended polychromate. The ordinary U-polynomial, Tutte's universal $V$-function, and Stanley's Tutte symmetric function are specializations of this refinement. We conjecture that L_G and U_G have the same distinguishing power on simple graphs, and verify this through eleven vertices. We exhibit five-vertex loopless multigraphs with equal U-polynomials but distinct loopy and extended U-polynomials, solving a problem of Merino and Noble. We also disprove Loebl's conjecture for chordal graphs and its restriction to split graphs, with counterexamples on eleven and fourteen vertices, respectively. For the score polytope P_G of the external bizonotopal algebra, we give a geometric realization of loopy deletion-contraction. The resulting forest-indexed geometric parking complexes partition the lattice points of P_G and are piecewise integral-affinely parametrized by boxes whose interval lengths are the component indices in the forest expansion.

Let $M$ be a left $R$-module. We define the \emph{homomorphism submodule graph} $Γ_{\mathrm{Hom}}(M)$ as the simple graph whose vertices are the proper submodules of $M$, with an edge between distinct vertices $N_1$ and $N_2$ if and only if $\mathrm{Hom}_R(N_1, M/N_2) \ne 0$ or $\mathrm{Hom}_R(N_2, M/N_1) \ne 0$. This graph encodes homological information about $M$ and reflects its internal structure. We compute $Γ_{\mathrm{Hom}}(M)$ for semisimple and uniserial modules, establish precise correspondences between graph-theoretic and algebraic properties, and prove that for modules over Artinian local rings, the isomorphism type of $M$ is determined by $Γ_{\mathrm{Hom}}(M)$. We also show that over commutative rings with identity, the graph is always chordal, and we relate its spectral radius to composition length in natural families.

Let $G$ be a finite simple graph and let $I_c(G)$ denote its complementary edge ideal. In this paper, we study the Castelnuovo-Mumford regularity, the depth and Serre depth for $(S_{2})$ of the symbolic powers of $I_c(G)$. We derive explicit formulas for $\operatorname{reg} I_c(G)^{(k)}$ and $\operatorname{reg} I_c(G)^{(k)}/I_c(G)^k$ for every $k\geq1$, in terms of the number and the sizes of the connected components of $G$. As applications, we determine their asymptotic behavior, and completely characterize the symbolic powers of complementary edge ideals having linear resolutions. We then investigate the symbolic depth function $k\mapsto\operatorname{depth} S/I_c(G)^{(k)}$. We determine explicitly its values for $k=2,3,4$, and establish bounds for the asymptotic depth. More generally, for an arbitrary squarefree monomial ideal $I$, we compare the local cohomology modules of $S/I^{(k)}$ for different symbolic powers, and derive corresponding inequalities for the depth and the Serre depth. Applying this to complementary edge ideals, we obtain $\operatorname{depth} S/I_{c}(G)^{(t)}\geq\operatorname{depth} S/I_{c}(G)^{(k)}$ for $k\geq 2t-1$, despite the fact that the symbolic depth function need not be non-increasing. Finally, we characterize complementary edge ideals satisfying Serre's condition $(S_r)$, and determine the Serre depth for $(S_{2})$ of all symbolic powers of complementary edge ideals. In particular, we obtain combinatorial characterizations of Serre's condition $(S_{2})$ and of the Cohen-Macaulayness of $S/I_{c}(G)^{(k)}$ for every $k\geq 1$.

We study the common spanning structure of connected finite simple graphs that are indistinguishable by uniformly initialized color refinement. We prove that any two such graphs with common cyclomatic number $r\ge2$ contain isomorphic spanning forests with at most $r-1$ components. This bound depends only on $r$ and is sharp for every $r\ge2$. We also determine the maximum, over all such graph pairs without restricting their order, of the minimum edge symmetric difference under vertex bijections: it is exactly $4(r-1)$.

The lattice-coset method of Alon, Bucić, and Davies gives real solutions with controlled coordinates. We combine it with integer rounding to prescribe outdegrees and flows from spanning-tree packing. For every integer $q\ge3$, $q$ edge-disjoint spanning trees suffice for every outdegree prescription modulo $q$ whose sum is compatible with the number of edges. The tree count is optimal even for simple graphs. A common interval theorem extends the argument to totally unimodular matrices; for graphs, its partition inequalities admit a sharp allowance of two. Fractional packing characterizes when every prescription has a strongly connected realization and gives circular-flow bounds. In particular, for every integer $k\ge1$, every $(4k+2)$-edge-connected graph has circular flow number strictly below $2+1/k$, with optimal connectivity for this strict inequality. As a structural application, every four-edge-connected graph with no bond of size six, seven or eight has a nowhere-zero three-flow. Further consequences concern star decompositions, antisymmetric flows and contraction.

Let $G=(V,E)$ be a simple graph and $(2k+1)$ be a prime integer. Let each vertex of $G$ be colored using one of the $(2k+1)$ colors, say $R_1,R_2,...,R_{2k+1}$. If every vertex has an equal number of neighbors of each color, then the coloring is a $(2k+1)$-neighborhood balanced coloring. We establish a number of results for common families of graphs and present some families of graphs that have this property.

Let $λ_1(n,Δ)$ be the maximum adjacency spectral radius among connected nonregular simple graphs of order $n$ and maximum degree $Δ$. Using effective resistance bounds, explicit comparison graphs and a one-dimensional Wirtinger inequality, we prove that, for every fixed integer $Δ\ge3$, $λ_1(n,Δ)=Δ-c_Δπ^2/(4n^2)+O_Δ(n^{-5/2})$ as $n\to\infty$, where $c_Δ=Δ-1$ for odd $Δ$ and $c_Δ=2(Δ-2)$ for even $Δ$. This proves the asymptotic conjecture posed by Liu [J. Combin. Theory Ser. B 169 (2024), Conjecture 7.1].

Razborov's flag algebra method is one of the most successful tools for proving asymptotic density inequalities in extremal graph theory. It reduces an inequality about all large graphs avoiding a forbidden pattern to finitely many algebraic conditions, and a certificate satisfying them can be found by semidefinite programming. We formalize the method for finite simple graphs in Lean 4, building on Mathlib. The formalization covers flags and their densities, the flag algebra as a quotient vector space, positive homomorphisms as limits of graph sequences, Razborov's random extensions, and the downward operator that averages out labels. On top of this theory, we build verified computation of flag densities and a tactic that turns certificates found by semidefinite programming into Lean proofs. The certificate serves only as candidate data; Lean checks every fact the proof uses. With this development we verify seven Turán-type upper bounds. Adding lower-bound constructions proved in Lean yields the asymptotic form of Mantel's theorem and the asymptotic Erdős pentagon theorem $π(C_5;K_3)=24/625$; to our knowledge, the latter has not previously been formalized in a proof assistant. We also prove two inequalities of Goodman. All density theorems reported here depend only on the standard axioms of Lean's logic. Formalization also led us to treat forbidden subgraphs with a different semantics. Instead of building a separate flag algebra for each forbidden family, we keep a single algebra of all graphs and impose the constraint only when interpreting an inequality, through randomly labeled limits of graphs satisfying it. We prove that this ensemble semantics yields the intended density bounds, and that a condition we call root-plantability characterizes when it agrees with the standard semantics.

We show that there exist graphons that interpolate between Riemannian manifolds and weighted geometric graphs. Specifically, the graph-to-manifold approximation used in manifold learning can be regarded as the composition of a graph-to-graphon convergence and a graphon-to-manifold convergence in a certain sense. Furthermore, we establish a monotonicity inequality which reveals an implicit relationship between numerous combinatorial parameters and geometric quantities on graphons. Using this inequality, we find relations among conductance, maxcut problem, capacity, and packing radius, as well as their limiting behaviors under graph-to-graphon and graphon-to-manifold convergences; some of these relations are novel even for simple graphs and closed manifolds. We also find that in the graphon-to-manifold limiting process, the purely combinatorial parameters diverge, leading to a ``blowup'' phenomenon; and continuous versions of some graph parameters cannot be realized via Riemannian manifolds, but can be realized via graphons.

In this paper, we introduce and investigate the $\textbf{k}$-neighborhood ideal of a graph, a natural generalization of the closed neighborhood ideal. Let $G$ be a simple graph on the vertex set $[n]$, and let $S=K[x_1,\dots,x_n]$ be the polynomial ring over a field $K$. For a vector $\textbf{k}=(k_1,\ldots,k_n)\in \mathbb{N}^n$ satisfying $1\leq k_i\leq \textrm{deg}_G(i)+1$ for all $i$, the $\textbf{k}$-neighborhood ideal of $G$ is defined as the squarefree monomial ideal $$\textrm{NI}_{\textbf{k}}(G)=\sum_{i=1}^n\, (\textbf{x}_W:\, W\subseteq N_G[i],\, |W|=k_i)$$ of $S$, where $\textbf{x}_W=\prod_{i\in W} x_i$. We study homological invariants and properties of $\textrm{NI}_{\textbf{k}}(G)$ focusing on its Castelnuovo-Mumford regularity, projective dimension and Cohen-Macaulayness. Special attention is devoted to the case where the vector ${\textbf{k}}$ is the degree-vector of the graph, i.e., $k_i=\textrm{deg}_G(i)$ for all vertices $i$, and to the case where $\textrm{NI}_{\textbf{k}}(G)$ coincides with the edge ideal of a graph. In these settings, we provide combinatorial characterizations and bounds for the regularity and projective dimension of $\textrm{NI}_{\textbf{k}}(G)$ for several classes of graphs, and further investigate the Cohen-Macaulay property of these ideals.