Largest symmetric intersecting families set-intersecting
Largest symmetric intersecting families are set-intersecting · arXiv:1909.11578
Status open high confidence
The conjecture that the largest symmetric intersecting families in $[k]^n$ are set-intersecting remains open. The source paper (published in Combinatorics, Probability and Computing 2021) established only the size bound $o(k^n)$ for fixed $k \geq 3$; the structural question about extremal families is explicitly flagged as potentially unprovable. A follow-up by Keevash (arXiv 2103.05050, JLMS 2023) reproves the size bound via forbidden intersections for codes, but no paper resolving the structural conjecture was found in an exhaustive web search through May 2026.
Reviewer notes. No follow-up resolving the structural conjecture was found in the indexed literature. The authors themselves hedge the conjecture as 'maybe impossible to prove', suggesting it is widely regarded as very hard. Keevash's 2021 forbidden-intersections-for-codes paper (arXiv 2103.05050) reproves the $o(k^n)$ size bound but does not address the set-intersecting structure question; the Wiley DOI returned a paywall error so it could not be fully verified and is not included as a cite.
Context
A family $A \subset [k]^n$ is set-intersecting if there exists a symmetric intersecting family $\mathcal{B}$ of subsets of $[n]$ and $\ell \in [k]$ such that $x \in A$ if and only if there is some $B \in \mathcal{B}$ with $x_i = \ell$ for all $i \in B$. The authors note that all their best constructions have this form, motivating the conjecture.
Notes. PDF source — math notation may be garbled. Stated with the explicit hedge 'seems possible (but maybe impossible to prove)'; no labeled environment.
Source paper
On symmetric intersecting families of vectors
Sean Eberhard, Jeff Kahn, Bhargav Narayanan, Sophie Spirkl · 2020-07-31
https://arxiv.org/abs/1909.11578
PDF source