{"id":"arxiv/2603.03086","url":"https://arxiv.org/abs/2603.03086","tags":["math.CO"],"score":0,"title":"Resolving the Kohayakawa--Kreuter Conjecture for Families","domain":"math.CO","source":"arxiv","authors":["Matthew Yancey"],"reviews":0,"subject":"math.CO","summary":"A graph $G$ is $(a,b)$-sparse if every nonempty subgraph $H$ satisfies $e(H) \\leq a v(H) - b$. We are interested in the conditions under which an $(a,b)$-sparse graph can be partitioned $E(G) = E(G_1) \\cup E(G_2)$ such that for $i \\in \\{1,2\\}$ we have that $G_i$ is $(a_i, b_i)$-sparse. Kuperwasser, Samotij, and Wigderson conjectured that a $(m,0)$-sparse graph can be partitioned into a $(1,1)$-sparse graph and a $(m,2m-1)$-sparse graph. We prove the conjecture in full. The Kohayakawa--Kreuter Conjecture for Families claims that $n^{-1/m_2}$ is the threshold function for the random graph…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-03-03","reactions":{}}