# Localization: A Framework to Generalize Extremal Graph Problems

## TL;DR

This paper introduces a localization framework to improve bounds in extremal graph theory and characterizes the structure of extremal graphs for various problems.

## Contribution

It develops a new localization approach that refines classical extremal bounds and provides structural insights into extremal graphs.

## Key findings

- Improved bounds for the number of cliques in graphs with degree constraints.
- Structural characterizations of extremal graphs achieving these bounds.
- Generalization of classical extremal results using localization.

## Abstract

Extremal graph theory studies the maximum or minimum number of subgraphs isomorphic to a prescribed graph under given constraints. \textit{Localization} has recently emerged as a framework that refines such problems by assigning extremal quantities locally (to vertices or edges) and then aggregating them. This perspective not only recovers classical results but also leads to sharper bounds. A classical result states that a connected planar graph with a finite girth $g$ satisfies \begin{equation*} m \leq \frac{g}{g-2}(n-2) \end{equation*} Wood~\cite{wood} derived upper bounds on the number of $K_t$-cliques in graphs of bounded maximum degree, expressed in terms of both the number of vertices and the number of edges: \begin{align*} ex(n,K_t,K_{1,d+1}) \leq \frac{n}{d+1}\binom{d+1}{t} \\ mex(m,K_t,K_{1,d+1}) \leq \frac{m}{\binom{d+1}{2}}\binom{d+1}{t} \end{align*} More recently, Chakraborty and Chen~\cite{CHAKRABORTI2024103955} established a similar upper bound for graphs with bounded path length: \begin{equation*} mex(m,K_t,P_{r+1}) \leq \frac{m}{\binom{r}{2}}\binom{r}{t} \end{equation*} In this paper, we employ the localization framework to improve these bounds and provide structural characterizations of the extremal graphs attaining them.