pith. sign in

arxiv: 2508.19723 · v2 · submitted 2025-08-27 · 🧮 math.CO

Two results on set families: sturdiness and intersection

Pith reviewed 2026-05-18 21:40 UTC · model grok-4.3

classification 🧮 math.CO
keywords extremal set theoryIU-familiessturdinessintersecting familiesseparated familiescross t-intersectingunion constraints
0
0 comments X

The pith

Every IU-family of subsets has sturdiness at most 2 to the power of n minus 4.

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

The paper proves that any family of subsets where every two sets share at least one element and no two together cover the whole ground set must have sturdiness at most 2^{n-4}. Sturdiness is the smallest number of sets that contain one chosen element but not another, after the first element is dropped. This bound confirms a conjecture by Frankl and Wang. The authors also determine the largest possible total size for a pair of cross t-intersecting separated families. They supply explicit constructions that show an earlier open question on such families has a negative answer.

Core claim

The paper shows that if F is an IU-family in the power set of an n-element set, then its sturdiness β(F) is at most 2^{n-4}. It further gives a tight upper bound on the sum of the sizes of two cross t-intersecting separated families and supplies explicit counterexamples to an open problem posed by Frankl, Liu, Wang and Yang.

What carries the argument

Sturdiness β(F), defined as the minimum over all i ≠ j of the size of F(i, bar j), the collection of sets in F that contain i but not j, each with i removed.

If this is right

  • The known size bound of 2^{n-2} for IU-families is now joined by a uniform lower bound on how many sets must be asymmetric with respect to every pair of elements.
  • The combined size of any two cross t-intersecting separated families is capped by a specific tight expression.
  • The open problem on separated families is settled negatively by the counterexamples given.

Where Pith is reading between the lines

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

  • The sturdiness bound may make it easier to classify the largest possible IU-families by ruling out overly lopsided distributions.
  • The counterexamples suggest that the extremal size for cross t-intersecting pairs can grow differently once t increases beyond 1.
  • Similar minimum-size tracking arguments could be applied to other pairs of constraints on set families.

Load-bearing premise

The minimum size of sets containing one element but not the other correctly reflects the global limits imposed by requiring every pair to intersect yet forbidding any pair from covering the entire ground set.

What would settle it

An explicit IU-family on n elements in which the smallest F(i, bar j) exceeds 2^{n-4} for some pair i and j would disprove the bound.

read the original abstract

This paper resolves two open problems in extremal set theory. For a family $\mathcal{F} \subseteq 2^{[n]}$ and $i, j\in [n]$, we denote $\mathcal{F} (i,\bar{j})=\{F\backslash\{i\}: F\in \mathcal{F}, F\cap\{i,j\}=\{i\}\}$. The sturdiness $\beta (\mathcal{F})$ is defined as the minimum $|\mathcal{F} (i,\bar{j})|$ over all $i\neq j$. A family $\mathcal{F}$ is called an IU-family if it satisfies the intersection constraint: $F\cap F'\neq \emptyset $ for all $F,F'\in \mathcal{F}$, as well as the union constraint: $F\cup F' \neq [n]$ for all $F,F'\in \mathcal{F}$. The well-known IU-Theorem states that every IU-family $\mathcal{F}\subseteq 2^{[n]}$ has size at most $ 2^{n-2}$. In this paper, we prove that if $\mathcal{F}\subseteq 2^{[n]}$ is an IU-family, then $\beta (\mathcal{F})\le 2^{n-4}$. This confirms a recent conjecture proposed by Frankl and Wang. As the second result, we establish a tight upper bound on the sum of sizes of cross $t$-intersecting separated families. Our result not only extends a previous theorem of Frankl, Liu, Wang and Yang on separated families, but also provides explicit counterexamples to an open problem proposed by them, thereby settling their problem in the negative.

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.

Referee Report

0 major / 3 minor

Summary. The paper resolves two open problems in extremal set theory. It proves that if F ⊆ 2^[n] is an IU-family (satisfying both the intersecting condition F ∩ F' ≠ ∅ and the union condition F ∪ F' ≠ [n] for all distinct F, F' in F), then the sturdiness β(F) — defined as the minimum of |F(i, ¯j)| over i ≠ j — satisfies β(F) ≤ 2^{n-4}, confirming a conjecture of Frankl and Wang. It also establishes a tight upper bound on the sum of sizes of cross t-intersecting separated families and supplies explicit counterexamples that settle in the negative an open problem posed by Frankl, Liu, Wang and Yang.

Significance. If the results hold, they advance extremal set theory by strengthening the classical IU-Theorem (|F| ≤ 2^{n-2}) with a uniform lower bound on the size of all slices F(i, ¯j), and by extending prior work on separated families with a sharp sum bound together with concrete constructions that falsify the open question. The combinatorial arguments via averaging over pairs and exhaustive case analysis on the slices, together with the explicit counterexamples, constitute verifiable contributions that can be checked directly against the definitions.

minor comments (3)
  1. [Abstract] The abstract introduces the notation F(i, ¯j) and the IU conditions but does not explicitly restate the precise definition of an IU-family before stating the new bound; adding one sentence would improve readability for readers who consult only the abstract.
  2. [Section on cross t-intersecting families] The claim that the bound on cross t-intersecting separated families is tight would be strengthened by an explicit reference to the section or construction that achieves equality (e.g., the extremal example for t=1).
  3. [Introduction] A short remark comparing the new 2^{n-4} bound with the classical 2^{n-2} size bound of the IU-Theorem would clarify the improvement for readers.

Simulated Author's Rebuttal

0 responses · 0 unresolved

We thank the referee for the positive report, accurate summary of our two main results, and recommendation of minor revision. We appreciate the recognition that our proofs and counterexamples advance the understanding of IU-families and separated families in extremal set theory.

Circularity Check

0 steps flagged

No significant circularity detected

full rationale

The paper derives its bounds through direct combinatorial case analysis on the slices F(i,¯j) and averaging arguments that apply the external IU-Theorem bound |F| ≤ 2^{n-2} as an independent input, without reducing the target inequality β(F) ≤ 2^{n-4} to a redefinition or fitted parameter. The second result supplies explicit constructions that falsify the prior open problem by direct verification, extending results from non-overlapping authors. No self-definitional steps, load-bearing self-citations, or ansatz smuggling appear in the derivation chain.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The central claims rest on standard definitions of the power set, intersection and union constraints, and the minimum over derived subfamilies; no free parameters or invented entities are introduced.

axioms (2)
  • standard math The power set 2^[n] with the standard subset, intersection, and union operations forms a Boolean lattice.
    Invoked implicitly when defining families F ⊆ 2^[n] and the derived sets F(i, j-bar).
  • domain assumption Every IU-family satisfies both the intersection constraint F ∩ F' ≠ ∅ and the union constraint F ∪ F' ≠ [n] for all distinct F, F'.
    This is the defining property used to derive the sturdiness bound.

pith-pipeline@v0.9.0 · 5830 in / 1503 out tokens · 32259 ms · 2026-05-18T21:40:35.939676+00:00 · methodology

discussion (0)

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

Reference graph

Works this paper leans on

24 extracted references · 24 canonical work pages

  1. [1]

    Anderson, Intersection theorems and a lemma of Kleitman, Discrete Math

    I. Anderson, Intersection theorems and a lemma of Kleitman, Discrete Math. 16 (3) (1976) 181–185

  2. [2]

    Daykin, L

    D. Daykin, L. Lov´ asz, The number of values of a Boolean function, J. Lond. Math. Soc. (2) 12 (1975/1976) 225–230

  3. [3]

    Erd˝ os, C

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

  4. [4]

    Frankl, The proof of a conjecture of G.O.H

    P. Frankl, The proof of a conjecture of G.O.H. Katona, J. Combin. Theory Ser. A 19 (2) (1975) 208–213

  5. [5]

    Frankl, Antichains of fixed diameter, Mosc

    P. Frankl, Antichains of fixed diameter, Mosc. J. Comb. Number Theory 7 (2017) 3–33

  6. [6]

    Frankl, Maximal degree and diversity in intersecting hypergraphs, J

    P. Frankl, Maximal degree and diversity in intersecting hypergraphs, J. Combin. Theory Ser. B 144 (2020) 81–94

  7. [7]

    Frankl, A

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

  8. [8]

    Frankl, E

    P. Frankl, E. Liu, J. Wang, Z. Yang, Some inequalities concerning cross-intersecting families of integer sequences, Discrete Math. 346 (2023) 113574

  9. [9]

    Frankl, E

    P. Frankl, E. Liu, J. Wang, Z. Yang, Non-trivial t-intersecting separated families, Discrete Appl. Math. 342 (2024) 124–137

  10. [10]

    Frankl, N

    P. Frankl, N. Tokushige, Invitation to intersection problems for finite sets, J. Combin. Theory Ser. A 144 (2016) 157–211

  11. [11]

    Frankl, J

    P. Frankl, J. Wang, Best possible bounds on the double-diversity of intersecting hyper- graphs, (2022), arXiv:2212.11650

  12. [12]

    Frankl, J

    P. Frankl, J. Wang, The maximum sturdiness of intersecting families, (2024), arXiv:2412.07090. 10

  13. [13]

    Frankl, J

    P. Frankl, J. Wang, Improved bounds on the maximum diversity of intersecting families, European J. Combin. 118 (2024) 103885

  14. [14]

    Frankl, J

    P. Frankl, J. Wang, On the C-diversity of intersecting hypergraphs, European J. Combin. 130 (2025) 104199

  15. [15]

    Gupta, Y

    P. Gupta, Y. Mogge, S. Piga, B. Schulke, r-cross t-intersecting families via necessary inter- section points, Bull. Lond. Math. Soc. 55 (3) (2023) 1447–1458

  16. [16]

    Huang, Two extremal problems on intersecting families, European J

    H. Huang, Two extremal problems on intersecting families, European J. Combin. 76 (2019) 1–9

  17. [17]

    Kleitman, Families of non-disjoint subsets, J

    D. Kleitman, Families of non-disjoint subsets, J. Combin. Theory 1 (1966) 153–155

  18. [18]

    Kupavskii, Diversity of uniform intersecting families, European J

    A. Kupavskii, Diversity of uniform intersecting families, European J. Combin. 74 (2018) 39–47

  19. [19]

    Lemons, C

    N. Lemons, C. Palmer, Unbalance of set systems, Graphs Combin. 24 (2008) 361–365

  20. [20]

    Liu, The maximum sum of the sizes of cross-intersecting separated families, AIMS Math

    E. Liu, The maximum sum of the sizes of cross-intersecting separated families, AIMS Math. 8 (12) (2023) 30910–30921

  21. [21]

    Marica, J

    J. Marica, J. Sch¨ onheim, Differences of sets and a problem of Graham, Canad. Math. Bull. 12 (1969) 635–637

  22. [22]

    Patk´ os, Size, diversity, minimum degree, sturdiness, d¨ omd¨ od¨ om, (2025), arXiv:2501.02596

    B. Patk´ os, Size, diversity, minimum degree, sturdiness, d¨ omd¨ od¨ om, (2025), arXiv:2501.02596

  23. [23]

    Seymour, On incomparable collections of sets, Mathematika, 20 (1973) 208–209

    P. Seymour, On incomparable collections of sets, Mathematika, 20 (1973) 208–209

  24. [24]

    Siggers, N

    M. Siggers, N. Tokushige, The maximum size of intersecting and union families of sets, European J. Combin. 33 (2012) 128–138. 11