Kernelization for -Packing Revisited
arXiv:2607.14779
The paper investigates kernelization for the H‑Packing problem, providing improved polynomial kernels for various subdivided‑star patterns and proving compression lower bounds for line patterns, showing that kernel size can increase when the pattern is slightly altered.
Abstract
\textsc{-Packing} asks whether a graph contains vertex-disjoint copies of a fixed pattern graph . Via the standard reduction to \textsc{-Set Packing}, one obtains generic kernels with vertices and edges. We revisit the question of beating these bounds for specific patterns . Our main results concern subdivided stars. Let denote the subdivided star with branches of length and branches of length . We obtain kernels with vertices and edges for , for , and for every , kernels with vertices and edges for every fixed with , and a kernel with vertices and edges for the paw. Our proofs proceed in two steps. First, we reduce to instances in which all but a small part of the graph is independent, or in which the graph has a small vertex cover. Second, we reduce the independent side by keeping only a bounded number of witness vertices for each subset of the small part. On the negative side, we prove a lower bound for the line . For every and every , \textsc{-Packing} does not admit a compression of size unless $\NP\subseteq \coNP/\poly$. Thus, deleting a single vertex from the pattern may, surprisingly, make kernelization provably harder, showing that compressibility of \textsc{-Packing} is not monotone under taking induced subgraphs.
ESA 2026