Pith. sign in

REVIEW 3 minor 2 cited by

Non-trivial cross-intersecting k-uniform families satisfy |A||B| <= h(n,k)^2 when n >= 2k and k >= 3.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Proves |A|*|B| <= h(n,k)^2 for non-trivial cross-intersecting k-uniform families A, B on [n] when n >= 2k and k >= 3, with extremal pair characterization.

T0 review reviewed 2026-06-26 challenge →

load-bearing objection This paper settles the Frankl-Wang conjecture on the sharp product bound for non-trivial cross-intersecting k-uniform families across the full range k >= 3 and n >= 2k.

arxiv 2606.23322 v1 pith:JCST764T submitted 2026-06-22 math.CO

A sharp product bound for non-trivial cross-intersecting families

Yang Huang This is my paper
classification math.CO MSC 05D05
keywords cross-intersecting familiesnon-trivial familiesHilton-Milner familyproduct boundextremal set theoryshift operationuniform hypergraphs
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

The reading

The paper proves that any two non-trivial cross-intersecting families of k-subsets of an n-set have size product at most h(n,k)^2 under the conditions n >= 2k and k >= 3. Here h(n,k) is the size of the Hilton-Milner family, given by binom(n-1,k-1) minus binom(n-k-1,k-1) plus one. This confirms the conjecture of Frankl and Wang in the full range and gives the exact equality cases. The result sharpens the earlier Pyber bound by ruling out the star families.

Core claim

Every non-trivial cross-intersecting pair A, B subset binom([n],k) with n >= 2k and k >= 3 satisfies |A||B| <= h(n,k)^2. Moreover, we characterize all extremal pairs. The product extremum forces a balanced, symmetric-or-dual structure: the two families are isomorphic when n>2k and complement-dual when n=2k.

What carries the argument

Diversity technique together with new properties of an extended shift operation.

Load-bearing premise

The diversity technique together with new properties of an extended shift operation suffices to cover all cases in the range n >= 2k and k >= 3.

What would settle it

A concrete pair of non-trivial cross-intersecting families A and B on [n] with n >= 2k, k >= 3 and |A||B| > h(n,k)^2 would falsify the bound.

Watch this falsifier. Get emailed when new claim-graph text bears on it.

If this is right

  • The product bound forces the two families to adopt a balanced symmetric-or-dual structure.
  • Extremal pairs are isomorphic for n > 2k and complement-dual for n = 2k.
  • The problem exhibits new extremal configurations that vary with uniformity.
  • A related conjecture of Frankl and Wang is disproved by the new configurations.

Where Pith is reading between the lines

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

  • The shift-based method may adapt to related intersection problems with different uniformity constraints.
  • The distinction between product and sum versions suggests that product bounds impose stricter symmetry than sum bounds do.
  • Partial results for small k or large n may now be unified under the same argument.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit.

Referee Report

0 major / 3 minor

Summary. The manuscript proves the Frankl-Wang conjecture: for non-trivial cross-intersecting k-uniform families A, B ⊆ binom([n],k) with n ≥ 2k and k ≥ 3, |A||B| ≤ h(n,k)^2 where h(n,k) = binom(n-1,k-1) - binom(n-k-1,k-1) + 1 is the Hilton-Milner size. It characterizes all extremal pairs (isomorphic when n > 2k; complement-dual when n = 2k). The proof combines the diversity technique with new properties of an extended shift operation. The paper also exhibits new extremal configurations for varying uniformities and disproves a related conjecture of Frankl and Wang. An independent contemporaneous proof covers the subrange k ≥ 8 and n ≥ 2k + 1.

Significance. Resolving the conjecture in the full stated range supplies the sharp product bound together with a complete extremal characterization that forces balanced symmetric-or-dual structure (in contrast to the asymmetric extremizers known for the sum version). The explicit disproof of the related conjecture and the identification of new extremal configurations for different k add concrete new information to the theory of cross-intersecting families. The combination of diversity arguments with refined shift operations, plus external corroboration via the independent proof, strengthens the result.

minor comments (3)
  1. The statement of the disproved related conjecture (mentioned in the abstract) should be recalled explicitly in the introduction or in a dedicated subsection so that readers can see the precise claim being refuted.
  2. Notation for the extended shift operation is introduced in the proof section; a short preliminary definition or diagram would improve readability before the technical lemmas are applied.
  3. The characterization theorem distinguishes the n = 2k and n > 2k cases; a single illustrative example for each case (with explicit families) would help readers verify the claimed isomorphism or duality.

Simulated Author's Rebuttal

0 responses · 0 unresolved

We thank the referee for their positive evaluation of our manuscript and for recommending acceptance. We appreciate the accurate summary of our results and the recognition of the significance of resolving the Frankl-Wang conjecture in the full range.

Circularity Check

0 steps flagged

Direct combinatorial proof; no circularity

full rationale

The paper establishes the Frankl-Wang conjecture via a direct argument combining the diversity technique with new properties of an extended shift operation. No parameters are fitted to data and then relabeled as predictions, no definitions are self-referential, and no load-bearing step reduces to a self-citation chain or imported uniqueness theorem. The extremal characterization follows from the same combinatorial reductions rather than being presupposed. The result is therefore self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The central claim rests on standard combinatorial axioms together with newly established properties of the shift operation; no free parameters or invented entities are introduced.

axioms (2)
  • standard math Binomial coefficient identities and basic properties of intersecting families
    Invoked throughout arguments on cross-intersecting pairs and Hilton-Milner size h(n,k).
  • domain assumption Hilton-Milner family achieves the maximum size among non-trivial intersecting families
    Used as the base quantity h(n,k) for the product bound.

reviewed 2026-06-26 · how reviews work

0 comments
Cite this review

Pith. "Pith review of A sharp product bound for non-trivial cross-intersecting families." pith.science (2026). https://pith.science/paper/JCST764T

@misc{pith2026260623322,
  author       = {Pith},
  title        = {Pith review of: A sharp product bound for non-trivial cross-intersecting families},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/JCST764T}},
  note         = {Machine review of arXiv:2606.23322}
}
Share X Bluesky LinkedIn Reddit HN
abstract

Two families $\mathcal{A}, \mathcal{B} \subset \binom{[n]}{k}$ are cross-intersecting if $A \cap B \ne \emptyset$ for all $A \in \mathcal{A}$ and $B \in \mathcal{B}$, and non-trivial if neither $\m A$ nor $\m B$ is a star. Pyber proved that any two cross-intersecting families $\mathcal{A}, \mathcal{B} \subset \binom{[n]}{k}$ satisfy $|\mathcal{A}||\mathcal{B}| \le \binom{n-1}{k-1}^2$, and the maximum is attained by two full stars. Frankl, as well as Frankl and Wang, conjectured that the sharp bound, when both families are required to be non-trivial, is $h(n,k)^2$, where $h(n,k) = \binom{n-1}{k-1} - \binom{n-k-1}{k-1} + 1$, the size of the Hilton--Milner family. The cases $k=3$, and the range $k\ge8$ and $n\ge 4k$, were established earlier by Frankl and by Frankl and Wang, respectively. In this paper, we prove their conjecture in the full range. We show that every non-trivial cross-intersecting pair $\mathcal{A}, \mathcal{B} \subset \binom{[n]}{k}$ with $n \ge 2k$ and $k \ge 3$ satisfies $|\mathcal{A}||\mathcal{B}| \le h(n,k)^2$. Moreover, we characterize all extremal pairs. Whereas the corresponding sum problem admits asymmetric and unbalanced extremizers, the product extremum forces a balanced, symmetric-or-dual structure: the two families are isomorphic when $n>2k$ and complement-dual when $n=2k$. Independently and contemporaneously with the present work, Frankl and Wang obtained the same bound for $k\ge8$ and $n\ge2k+1$ by a different method. Our proof combines a diversity technique with several new properties of an extended shift operation. Moreover, we show that the problem behaves differently for different uniformities, exhibiting new extremal configurations. In particular, we disprove a related conjecture proposed by Frankl and Wang.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 2 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. On a conjecture regarding the product version of the Hilton-Milner theorem

    math.CO 2026-07 accept novelty 7.0

    A two-center construction falsifies the Frankl-Wang product Hilton-Milner conjecture for n up to (c_ℓ-ε)k, while the conjecture holds with full extremal characterization for n>100ℓk^{2}.

  2. On a conjecture regarding the product version of the Hilton-Milner theorem

    math.CO 2026-07 accept novelty 6.0

    The Frankl-Wang product version of the Hilton-Milner conjecture is false in a linear parameter range (via a two-center construction) but true for n > 100ℓk² with characterized extremal families.

Reference graph

Works this paper leans on

17 extracted references · 3 canonical work pages · cited by 1 Pith paper · 2 internal anchors

  1. [1]

    Daykin, Erd˝ os–Ko–Rado from Kruskal–Katona, Journal of Combinatorial Theory, Series A 17 (1974), 254–255

    D.E. Daykin, Erd˝ os–Ko–Rado from Kruskal–Katona, Journal of Combinatorial Theory, Series A 17 (1974), 254–255

    1974

  2. [2]

    Erd˝ os, C

    P. Erd˝ os, C. Ko, R. Rado, Intersection theorems for systems of finite sets, Quart. J. Math. Oxf. 2(12) (1961) 313–320

    1961

  3. [3]

    Frankl, On the maximum of the sum of the sizes of non-trivial cross-intersecting families, Combinatorica 44 (2024), 15–35

    P. Frankl, On the maximum of the sum of the sizes of non-trivial cross-intersecting families, Combinatorica 44 (2024), 15–35

    2024

  4. [4]

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

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

    2026

  5. [5]

    Frankl, Z

    P. Frankl, Z. F¨ uredi, Non-trivial intersecting families, Journal of Combinatorial Theory, Series A 41 (1986), 150–153

    1986

  6. [6]

    Frankl, A

    P. Frankl, A. Kupavskii, A size-sensitive inequality for cross-intersecting families, European Journal of Combinatorics, 62 (2017) 263–271

    2017

  7. [7]

    Frankl, A

    P. Frankl, A. Kupavskii, Diversity, Journal of Combinatorial Theory, Series A 182 (2021) 105468

    2021

  8. [8]

    Frankl, N

    P. Frankl, N. Tokushige, Some best possible inequalities concerning cross-intersecting families, Journal of Combinatorial Theory, Series A 61 (1992), N1, 87–97

    1992

  9. [9]

    Frankl, J

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

    2023

  10. [10]

    A product version of the Hilton-Milner Theorem II

    P. Frankl, J. Wang, A product version of the Hilton–Milner Theorem II, (2026), arXiv:2605.09246

    work page internal anchor Pith review Pith/arXiv arXiv 2026

  11. [11]

    Hilton, An intersection theorem for a collection of families of subsets of a finite set, Journal of the London Mathematical Society, 2 (1977) 369–376

    A.J.W. Hilton, An intersection theorem for a collection of families of subsets of a finite set, Journal of the London Mathematical Society, 2 (1977) 369–376

    1977

  12. [12]

    Hilton, The Erd˝ os–Ko–Rado theorem with valency conditions, 1976, unpublished manuscript

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

    1976

  13. [13]

    Hilton, E.C

    A.J.W. Hilton, E.C. Milner, Some intersection theorems for systems of finite sets, Quart. J. Math. Oxford Ser. (2) 18 (1967) 369–384

    1967

  14. [14]

    Structure and properties of large cross-intersecting families

    Y. Huang, A. Kupavskii, Structure and properties of large cross-intersecting families, (2026), arXiv:2606.20085

    work page internal anchor Pith review Pith/arXiv arXiv 2026

  15. [15]

    Huang, Y

    Y. Huang, Y. Peng, Complete characterization on maximum pairwise cross intersecting families (I), (2026), arXiv:2601.01929

    work page arXiv 2026

  16. [16]

    Matsumoto, N

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

    1989

  17. [17]

    Pyber, A new generalization of the Erd˝ os–Ko–Rado theorem, Journal of Combinatorial Theory, Series A, 43 (1986) 85–90

    L. Pyber, A new generalization of the Erd˝ os–Ko–Rado theorem, Journal of Combinatorial Theory, Series A, 43 (1986) 85–90. 30 Appendix In this section, we give the proof of Proposition 7.2. We need the following two lemmas. Lemma 8.3.Letn≥7, and letA,B ⊂ [n] 3 be non-trivial cross-intersecting families. If|A|= |B|= 3n−8, then∆(A)≥n−1or∆(B)≥n−1. Proof.Supp...

    1986

This paper was first reviewed by grok-4.3 on June 26, 2026.