arxiv: 2605.09246 · v1 · submitted 2026-05-10 · 🧮 math.CO

Recognition: no theorem link

A product version of the Hilton-Milner Theorem II

Jian Wang, Peter Frankl

Pith reviewed 2026-05-12 03:41 UTC · model grok-4.3

classification 🧮 math.CO

keywords cross-intersecting familiesHilton-Milner theoremextremal set theoryuniform intersecting familiesproduct boundsk-subsetsnon-trivial intersecting families

0 comments

The pith

Non-trivial cross-intersecting families of k-subsets have size product at most the square of the Hilton-Milner number when k is at least 8.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper proves that if two families of k-element subsets of an n-element set are cross-intersecting, with no single element common to all members of either family, then the product of their sizes is bounded above by the square of a specific number. This number equals the maximum size of a single non-trivial intersecting family as given by the Hilton-Milner theorem. The inequality holds whenever n is at least 2k plus one and k is at least 8. A reader cares because the result gives the precise maximum product for pairs of families in this classical setting of extremal set theory, extending the single-family case directly to products.

Core claim

If F and G are non-trivial cross-intersecting subsets of the k-subsets of [n], with n at least 2k+1 and k at least 8, then the product of their cardinalities is at most the square of (the binomial coefficient of n-1 choose k-1 minus the binomial coefficient of n-k-1 choose k-1 plus one).

What carries the argument

The non-trivial cross-intersecting condition on a pair of families, which requires every member of one to intersect every member of the other while each family itself has empty total intersection, combined with the product of the Hilton-Milner extremal size.

If this is right

The product of sizes for any such pair of families cannot exceed the stated quantity.
The bound completes the product version of the Hilton-Milner theorem in the full range n at least 2k+1 once k reaches 8.
Equality cases are controlled by the same extremal constructions that achieve the single-family Hilton-Milner bound.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

A separate argument might establish the same product bound for k between 2 and 7.
The result could be applied to bound the maximum size of a collection of several mutually cross-intersecting families.
Similar product bounds might hold when the families are required to be t-intersecting for t greater than 1.

Load-bearing premise

The claimed inequality holds only under the size restriction k at least 8, so the result depends on this threshold where the proof technique applies.

What would settle it

Two explicit non-trivial cross-intersecting families F and G of 8-subsets of a 17-element set whose size product exceeds the square of (binom(16,7) minus binom(9,7) plus one) would disprove the central claim.

read the original abstract

Two families $\mathcal{F},\mathcal{G}$ of $k$-subsets of $\{1,2,\ldots,n\}$ are called {\it non-trivial cross-intersecting} if $F\cap G\neq \emptyset$ for all $F\in \mathcal{F}, G\in \mathcal{G}$ and $\cap \{F\colon F\in \mathcal{F}\}=\emptyset=\cap \{G\colon G\in\mathcal{G}\}$. In this note, we establish the product version of the Hilton-Milner Theorem for $k\geq 8$ in the full range. That is, if $\mathcal{F},\mathcal{G}\subset \binom{[n]}{k}$ are non-trivial cross-intersecting, $n\geq 2k+1$ and $k\geq 8$, then \[ |\mathcal{F}||\mathcal{G}|\leq \left(\binom{n-1}{k-1}- \binom{n-k-1}{k-1} +1\right)^2. \]

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

Frankl and Wang close the product Hilton-Milner gap for k at least 8 with a sharp bound on cross-intersecting pairs.

read the letter

The punchline is that this paper completes the product version of the Hilton-Milner theorem for k at least 8 in the full range of n. For non-trivial cross-intersecting k-uniform families F and G, the product of their sizes is bounded by the square of the classical Hilton-Milner quantity. What stands out is the clean statement of the result. The bound matches exactly what one gets by squaring the single-family Hilton-Milner number, and the authors show it holds under the stated conditions. This extends prior product versions that were limited to smaller k or narrower parameter ranges. The work builds directly on the intersecting family literature without introducing new parameters or fitted constants. The main soft spot is the explicit cutoff at k equals 8. The proof technique apparently needs this size to handle the intersection cases without additional restrictions, leaving smaller k for future work. There is no indication of gaps in the argument for the covered range, and the bound is achieved by taking both families to be the standard Hilton-Milner example. The citation pattern looks standard for this area, drawing on the original Hilton-Milner paper and recent product versions. This paper is for researchers in extremal combinatorics who work with intersecting and cross-intersecting families. Anyone needing a sharp product bound in this setting will get direct value from it. It is not a foundational shift but a solid completion of a specific open case. I would bring this to a reading group focused on set theory extremal problems. The paper shows clear thinking on the problem and deserves a serious referee. I recommend sending it to peer review.

Referee Report

0 major / 3 minor

Summary. The paper proves a product version of the Hilton-Milner theorem: if F and G are non-trivial cross-intersecting k-uniform families on [n] with n ≥ 2k+1 and k ≥ 8, then |F| |G| ≤ [binom(n-1,k-1) - binom(n-k-1,k-1) + 1]^2.

Significance. If the result holds, it supplies the sharp product bound for non-trivial cross-intersecting pairs in the full range n ≥ 2k+1 when k ≥ 8, matching the square of the classical Hilton-Milner number and confirming that the extremal configuration is attained by taking both families to be the Hilton-Milner example. This completes the product analogue begun in Part I and strengthens the understanding of stability phenomena in intersecting set systems.

minor comments (3)

The abstract states the theorem cleanly but omits an explicit remark that the bound is tight; adding one sentence would improve readability.
Section 1 (Introduction) refers to the result as 'Part II' but does not include a citation or arXiv link to Part I; this should be added for completeness.
In the statement of the main theorem (presumably Theorem 1.1 or 2.1), the condition k ≥ 8 is listed alongside n ≥ 2k+1; a brief parenthetical note explaining why the case k < 8 is deferred would help readers.

Simulated Author's Rebuttal

0 responses · 0 unresolved

We thank the referee for the positive assessment of our paper and the recommendation for minor revision. The referee's summary correctly captures the main contribution: the sharp product bound for non-trivial cross-intersecting k-uniform families when k ≥ 8 and n ≥ 2k+1, completing the product analogue of the Hilton-Milner theorem begun in Part I.

Circularity Check

0 steps flagged

No significant circularity in derivation chain

full rationale

The paper states a direct upper bound on |F||G| for non-trivial cross-intersecting families that is exactly the square of the classical Hilton-Milner extremal number binom(n-1,k-1) - binom(n-k-1,k-1) + 1. This bound is presented as the theorem statement itself for the given parameter range k ≥ 8 and n ≥ 2k+1, with no equations that redefine the target quantity in terms of itself, no fitted parameters relabeled as predictions, and no load-bearing steps that reduce to unverified self-citations. The derivation therefore remains self-contained against the external benchmark of the known Hilton-Milner theorem without internal reduction to its own inputs.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The result rests on standard facts from extremal set theory such as the Erdős–Ko–Rado theorem and the original Hilton-Milner theorem; no new free parameters, ad-hoc axioms, or invented entities are introduced in the abstract statement.

axioms (1)

standard math Standard properties of intersecting families and binomial coefficient identities used in extremal combinatorics
Invoked implicitly to bound the sizes of non-trivial intersecting families

pith-pipeline@v0.9.0 · 5465 in / 1208 out tokens · 34506 ms · 2026-05-12T03:41:35.008902+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

17 extracted references · 17 canonical work pages

[1]

Erd˝ os, C

P. Erd˝ os, C. Ko, R. Rado, Intersection theorems for systems of finite sets, Quart. J. Math. Oxford Ser. 12 (1961), 313–320

work page 1961
[2]

Frankl, The shifting technique in extremal set theory, Surveys in Combinatorics 123 (1987), 81–110

P. Frankl, The shifting technique in extremal set theory, Surveys in Combinatorics 123 (1987), 81–110

work page 1987
[3]

Frankl, A simple proof of the Hilton-Milner theorem, Moscow J

P. Frankl, A simple proof of the Hilton-Milner theorem, Moscow J. Comb. Number Theory 8 (2019), 97–101

work page 2019
[4]

Frankl, The maximum of the product of non-trivial cross-intersecting 3-graphs, Combinatorics and Number Theory 15(1)(2026) 1–8

P. Frankl, The maximum of the product of non-trivial cross-intersecting 3-graphs, Combinatorics and Number Theory 15(1)(2026) 1–8

work page 2026
[5]

Frankl, Z

P. Frankl, Z. F¨ uredi, Non-trivial intersecting families, J. Combin. Theory Ser. A 41 (1986), 150–153

work page 1986
[6]

Frankl, A

P. Frankl, A. Kupavskii, Diversity, J. Comb. Theory, Ser. A 182 (2021), 105468

work page 2021
[7]

Frankl, N

P. Frankl, N. Tokushige, Some best possible inequalities concerning cross-intersecting families, J. Combin. Theory Ser. A 61 (1992), 87–97

work page 1992
[8]

Frankl, J

P. Frankl, J. Wang, A product version of the Hilton–Milner Theorem, J. Combin. Theory Ser. A 200 (2023) 105791

work page 2023
[9]

Hilton, The Erd˝ os-Ko-Rado Theorem with valency conditions, unpublished manuscript, 1976

A.J.W. Hilton, The Erd˝ os-Ko-Rado Theorem with valency conditions, unpublished manuscript, 1976

work page 1976
[10]

Hilton, E.C

A.J.W. Hilton, E.C. Milner, Some intersection theorems for systems of finite sets, Quart.J. Math. Oxford Ser. 18 (1967), 369–384

work page 1967
[11]

Hurlbert, V

G. Hurlbert, V. Kamat, New injective proofs of the Erd˝ os-Ko-Rado and Hilton-Milner theorems, Discrete Math. 341 (2018), 1749–1754

work page 2018
[12]

Kupavskii, D

A. Kupavskii, D. Zakharov, Regular bipartite graphs and intersecting families, J. Comb. Theory Ser. A 155 (2018), 180–189

work page 2018
[13]

Matsumoto, N

M. Matsumoto, N. Tokushige, The exact bound in the Erd˝ os-Ko-Rado theorem for cross-intersecting families, J. Combin. Theory Ser. A 52 (1989), 90–97

work page 1989
[14]

M¨ ors, A generalization of a theorem of Kruskal, Graphs Combin

M. M¨ ors, A generalization of a theorem of Kruskal, Graphs Combin. 1 (1985), 167– 183. 8

work page 1985
[15]

Pyber, A new generalization of the Erd˝ os-Ko-Rado theorem, J

L. Pyber, A new generalization of the Erd˝ os-Ko-Rado theorem, J. Combin. Theory, Ser. A 43 (1986), 85–90. A Proofs of Lemma 2.9 and Lemma 2.10. Proof of Lemma 2.9.Note that n−2 k−2 + n−4 k−2 = k−1 n−1 + (k−1)(n−k)(n−k−1) (n−1)(n−2)(n−3) n−1 k−1 . Letx= k−1 n−1 + (k−1)(n−k)(n−k−1) (n−1)(n−2)(n−3) . Dividing both sides by n−1 k−1 , we see that (2.8) is equ...

work page 1986
[16]

Fork= 8,9,10 and 2k+ 1≤n≤4k, one can check by direct computation that (2.8) holds

Thus, to show (A.1) it suffices to show that 5 7 k−1 < 1 32, which is true fork≥11. Fork= 8,9,10 and 2k+ 1≤n≤4k, one can check by direct computation that (2.8) holds. Proof of Lemma 2.10.Note that n k−2 = (k−1)n (n−k+ 2)(n−k+ 1) n−1 k−1 . Letx= (k−1)n (n−k+2)(n−k+1) . Dividing both sides by n−1 k−1 , we see that (2.9) is equivalent to x+ h(n, k)2/ n−1 k−1...

work page
[17]

Fork= 8,9,10 and 3k≤n≤4k, one can check by direct computation that (2.9) holds

Thus, to show (A.2) it suffices to show that 5 7 k−1 < 1 32, which is true fork≥11. Fork= 8,9,10 and 3k≤n≤4k, one can check by direct computation that (2.9) holds. 10

work page