Indexed metadata

Ordinary lines, point degrees, and matroid density

Matthew E. Kroeker, Ben Lund

Source record

Source: arXiv

Published: Oct 15, 2026

arXiv: 2610.21415

Open original source ↗

Source abstract

For every set of nn points in Cd\mathbb{C}^d with full affine span, we prove that some point lies on at least n+1−2n/(d+1)n+1-2n/(d+1) spanned lines, which is best possible in every dimension. By successive contraction, this implies the sharp bound r+(ℓ−3)(r2)r+(\ell-3)\binom r2 on the number of elements of a simple rank-rr complex-representable matroid excluding U2,ℓU_{2,\ell} as a minor, which removes the sufficiently-large-rank assumption from the bound of Geelen, Nelson, and Walsh. An arithmetic application shows that rank-rr integer matrices whose maximal minors have absolute value at most ΔΔ have at most O(Δr2)O(Δr^2) distinct columns, resolving a question of Averkov and Schymura. We prove that a set of nn points in C3\mathbb{C}^3 determines at least n2/40n^2/40 ordinary lines whenever no plane contains more than 2n/32n/3 of the points. This answers a question of Basit, Dvir, Saraf, and Wolf. We also prove upper bounds on average line length and lower bounds on the number of points incident to many spanned lines in high-dimensional point sets. The proofs combine contraction of concentrated flats with normalization of the point vectors and second- and fourth-moment estimates on their pairwise correlations. This approach builds on the design matrix method.

Evidence graph

No public relationships recorded yet.

Integrity note: This page is a factual metadata record created by deterministic ingestion. It is not a claim that the work moves a mathematical frontier or has been independently verified.

Ordinary lines, point degrees, and matroid density — Mathematical Frontier Network