OctoInformBeta

OpenAI preprint, September 2026 · Section 3. From cross mass to triangle mass

The cross-mass property: Vertex masses and a growth function determine the allowed weighted adjacency

A uniform condition on cross mass.

Hover over a blue or green term in the text, or a node in the graph, to find it in the other. Select it to read its declaration both in Lean and in words.

The Setting

Cross-mass bounds constrain weighted adjacency between vertex sets whose total weights lie below a prescribed threshold. Let GG be a simple graph on a vertex set VV, and let w:V→ℝw\colon V\to\mathbb R be a real weight function.

The vertex mass w(A)w(A) of a finite set A⊆VA\subseteq V is the sum in (1): w(A)=∑v∈Aw(v).w(A)=\sum_{v\in A}w(v).(1)✦ The cross edge mass ew(A,B)e_w(A,B) between finite sets A,B⊆VA,B\subseteq V is the sum over ordered adjacent pairs in (2); the sets may overlap: ew(A,B)=∑u∈A∑v∈Buv∈E(G)w(u)w(v).e_w(A,B)=\sum_{u\in A}\ \sum_{\substack{v\in B\\uv\in E(G)}}w(u)w(v).(2)✦

For a real parameter xx, the growth function gxg_x of the variable ss is given by (3), with gx(0)=0g_x(0)=0: gx(s)=s(x+max{0,log1s}).g_x(s)=s\left(x+\max\left\{0,\log\frac1s\right\}\right).(3)✦ Here log\log denotes the natural logarithm.

The Definition

Definition 1. A weighted graph (G,w)(G,w) has the cross-mass property with constant C∈ℝC\in\mathbb R if, for every real x≥1x\ge1 and every pair of finite sets A,B⊆VA,B\subseteq V with w(A)≤exw(A)\le e^x and w(B)≤exw(B)\le e^x, the bound (4) holds: ew(A,B)≤Cgx(w(A)).e_w(A,B)\le C\,g_x\bigl(w(A)\bigr).(4)✦

In the paper, the result reads: a weighted graph (G,w)(G,w) satisfies the cross-mass bound with constant CC if, for every x≥1x\ge1 and A,B⊆VA,B\subseteq V, the inequalities w(A),w(B)≤exw(A),w(B)\le e^x imply ew(A,B)≤Cgx(w(A))e_w(A,B)\le C\,g_x(w(A)).

What Is Checked

The definition is stated over a simple graph, a real vertex weight function and a real constant, with vertex mass and cross edge mass evaluated on finite vertex sets.