Sunflower Property in Extremal Combinatorics

• The sunflower property is defined for set systems where every pair of sets intersects in a common kernel while the remaining parts (petals) remain disjoint.
• Research has refined the classical Erdős–Rado thresholds using advanced methods like tensor slice-rank techniques, leading to exponentially improved bounds.
• Extensions of the property apply to k-uniform hypergraphs and finite vector spaces, highlighting distinct notions (general-position vs. set-like) and impacting anti-Ramsey and forbidden configuration problems.

Searching arXiv for recent and foundational papers on the sunflower property in extremal combinatorics, including set-system, restricted-intersection, random-process, and finite-vector-space variants. In extremal combinatorics, the sunflower property is the condition that a family of sets has a common pairwise intersection. More precisely, a family $S_1,\dots,S_r$ is an $r$-sunflower, or $\Delta$-system, if there exists a set $K$ such that $S_i\cap S_j=K$ for all $i\neq j$; $K$ is the core or kernel, and the sets $S_i\setminus K$ are the petals. An equivalent description is that every ground-set element belongs to no set, every set, or exactly one set (Chizewer, 2023). The sunflower property is the organizing notion behind the Erdős–Rado sunflower problem, its modern tensor- and regularity-based refinements, its finite-vector-space analogues, and several related forbidden-configuration problems (Fukuyama, 2014, Otal, 12 May 2026).

1. Definition and basic structural features

For set systems, the sunflower condition is

$S_i\cap S_j=K \qquad \text{for all } i\neq j.$

The common intersection $K$ is the core or kernel, and the petals are the differences $r$0. In actual set systems, if all pairwise intersections equal $r$1, then automatically every higher intersection also equals $r$2, and there is no extra analogue of a “general position” condition to impose (Otal, 12 May 2026).

This formulation is stronger than a mere statement about repeated intersections. The kernel determines a canonical decomposition of each member into a shared part and a petal, and the petals are pairwise disjoint. The empty-kernel case is especially important: a matching is a sunflower with empty core. At the opposite extreme, a family of identical sets is a sunflower whose kernel is the set itself; in extremal questions one therefore works with distinct sets or distinct members of a family, depending on context (Bennett et al., 19 Sep 2025).

The same pattern appears in several neighboring settings. In complete $r$3-uniform hypergraphs, an $r$4-sunflower is a family of edges whose intersection has at least $r$5 elements; this variant underlies anti-Ramsey results for rainbow subhypergraphs (Martínez-Sandoval et al., 2015). In the Duke–Erdős problem, the forbidden configuration is a sunflower with $r$6 petals and core of size exactly $r$7, which interpolates between matching problems and restricted-intersection problems (Kupavskii et al., 21 Nov 2025).

2. Classical extremal problem and its quantitative refinements

The classical Erdős–Rado sunflower lemma states that a family $r$8 of sets, each of cardinality at most $r$9, contains a sunflower of cardinality $\Delta$0 whenever

$\Delta$1

In $\Delta$2-uniform notation, one may equivalently define $\Delta$3 as the minimum integer $\Delta$4 such that every $\Delta$5-uniform family of size at least $\Delta$6 contains a sunflower of size $\Delta$7; then Erdős and Rado proved

$\Delta$8

and also a lower bound of order

$\Delta$9

Their conjecture is that for every $K$0, there is a constant $K$1 such that

$K$2

(Mishra, 1 Jun 2026).

A first general asymptotic improvement over the Erdős–Rado threshold was proved by Fukuyama, who showed that there exists an absolute constant $K$3 such that

$K$4

forces a sunflower of cardinality $K$5. In particular, when $K$6 for fixed $K$7, the threshold becomes

$K$8

which is exponentially smaller than the classical bound (Fukuyama, 2014).

Subsequent work continued to reduce the base of the exponential. A 2025 result established that a family $K$9 of $S_i\cap S_j=K$0-element sets contains a $S_i\cap S_j=K$1-sunflower if

$S_i\cap S_j=K$2

for some absolute constant $S_i\cap S_j=K$3, replacing a logarithmic base by a sub-logarithmic base for fixed $S_i\cap S_j=K$4 (Fukuyama, 21 Oct 2025). A separate 2026 paper claims a proof of the Erdős–Rado conjecture and states the explicit theorem

$S_i\cap S_j=K$5

which is of the conjectured form $S_i\cap S_j=K$6; this is best described as a claim rather than a settled theorem in the literature (Mishra, 1 Jun 2026).

3. Nonuniform sunflower-free families and tensor methods

A complementary line of work studies the nonuniform problem in the Boolean lattice $S_i\cap S_j=K$7, where the question is how large a sunflower-free family $S_i\cap S_j=K$8 can be. Naslund and Sawin proved that any sunflower-free family satisfies

$S_i\cap S_j=K$9

and hence

$i\neq j$0

Their argument uses the slice-rank method on the tensor

$i\neq j$1

which becomes diagonal after restricting to uniform layers (Naslund et al., 2016).

A 2026 improvement keeps the same exponential constant but improves the polynomial prefactor. It proves that any sunflower-free family $i\neq j$2 satisfies

$i\neq j$3

improving the earlier

$i\neq j$4

The technical innovation is a slice-rank lemma for $i\neq j$5-triangular tensors: if $i\neq j$6 is $i\neq j$7-triangular with nonzero diagonal entries, then $i\neq j$8. This makes it possible to work with a single global tensor ordered by set size rather than summing diagonal bounds over all uniform layers (Ahmadi et al., 29 Jun 2026).

These nonuniform bounds are often summarized in terms of the sunflower-free capacity

$i\neq j$9

for which the slice-rank approach gives

$K$0

(Naslund et al., 2016). A number-theoretic reformulation from 2025 connects the general fixed-$K$1 capacity

$K$2

to harmonic LCM-free sets and proves

$K$3

together with the equivalence

$K$4

(Tang et al., 23 Dec 2025).

4. Restricted intersections and forbidden-core variants

The sunflower property becomes more rigid when pairwise intersections are restricted in advance. If $K$5 is $K$6-intersecting, meaning that $K$7 for all distinct $K$8, then one can force sunflowers with significantly smaller families than in the unrestricted case. For $K$9-intersecting families, where $S_i\setminus K$0, it was proved that there exists an absolute constant $S_i\setminus K$1 such that every $S_i\setminus K$2-uniform $S_i\setminus K$3-intersecting family contains an $S_i\setminus K$4-sunflower whenever

$S_i\setminus K$5

As a consequence, for any $S_i\setminus K$6, there exists $S_i\setminus K$7 such that every $S_i\setminus K$8-uniform $S_i\setminus K$9-sunflower-free family of size $S_i\cap S_j=K \qquad \text{for all } i\neq j.$0 contains two sets whose intersection has size at least $S_i\cap S_j=K \qquad \text{for all } i\neq j.$1 (Chizewer, 2023).

A different refinement is the Duke–Erdős problem, which asks for the largest family $S_i\cap S_j=K \qquad \text{for all } i\neq j.$2 containing no sunflower with $S_i\cap S_j=K \qquad \text{for all } i\neq j.$3 petals and core of size exactly $S_i\cap S_j=K \qquad \text{for all } i\neq j.$4. Frankl and Füredi showed that for fixed $S_i\cap S_j=K \qquad \text{for all } i\neq j.$5, $S_i\cap S_j=K \qquad \text{for all } i\neq j.$6, and $S_i\cap S_j=K \qquad \text{for all } i\neq j.$7,

$S_i\cap S_j=K \qquad \text{for all } i\neq j.$8

where $S_i\cap S_j=K \qquad \text{for all } i\neq j.$9 is the largest size of a $K$0-uniform sunflower-free family. A 2024 extension pushes this asymptotic picture to the regime $K$1 with $K$2 polynomial in $K$3 and $K$4, and also proves stronger results for forbidden sunflowers with core at most $K$5 (Kupavskii et al., 2024).

In the case $K$6, odd $K$7, $K$8, and $K$9 sufficiently large, the extremal structure is known exactly. If $r$00 is the disjoint union of two cliques of size $r$01, then the extremal family is

$r$02

which lifts the exact graph extremal structure for $r$03 to the forbidden-sunflower setting (Kupavskii et al., 21 Nov 2025).

5. Finite-vector-space analogues and the split between two sunflower notions

For subspaces over finite fields, the sunflower property admits two natural analogues. Let $r$04 be the ambient vector space, and let a $r$05-space mean a $r$06-dimensional subspace. In the stronger, general-position notion, $r$07-spaces $r$08 form a sunflower with kernel $r$09 if

$r$10

and in addition the quotient petals $r$11 are in general position, equivalently

$r$12

where $r$13. In this setting, for $r$14 and $r$15, every $r$16-sunflower-free family $r$17 of $r$18-spaces satisfies

$r$19

and there are constructions of size

$r$20

for $r$21, based on iterated lifted MRD codes (Ihringer et al., 6 May 2025).

A 2026 note isolates the weaker notion that is most faithful to set systems. There, a set-like $r$22-sunflower is defined only by the pairwise-intersection condition

$r$23

with no condition on the span. The distinction is substantive: the constructions of Ihringer–Kupavskii for the general-position notion do not remain sunflower-free in the set-like sense. The note gives an explicit example inside a family of $r$24-spaces in $r$25, where three distinct $r$26-spaces $r$27 satisfy

$r$28

with

$r$29

so they form a set-like $r$30-sunflower but not a general-position sunflower (Otal, 12 May 2026).

The same paper also gives the first systematic construction tailored to the set-like problem. Let $r$31, $r$32, let $r$33 be the matrix representation of $r$34 over $r$35, and define

$r$36

Then

$r$37

is a set-like $r$38-sunflower-free family of $r$39-spaces of size

$r$40

for every $r$41. The mechanism is that for distinct $r$42,

$r$43

so pairwise intersections are one dimension larger than triple intersections, which excludes a set-like sunflower (Otal, 12 May 2026).

6. Methods, random processes, and structural interpretations

Modern work on the sunflower property is methodologically diverse. One influential line passes through regularity and Boolean complexity. A 2019 paper introduced a structure-vs-pseudorandomness framework for set systems, defining $r$44-regularity by the condition

$r$45

for every $r$46, and showed that improved monotone DNF compression would imply improved sunflower theorems. Under a weaker upper-bound compression conjecture, it derives that for every $r$47 there exists $r$48 such that any $r$49-set system of size

$r$50

contains an $r$51-sunflower (Lovett et al., 2019).

A different regularity condition was developed for the 3-petal problem through the $r$52-condition

$r$53

for every nonempty $r$54. For each $r$55, there exists $r$56 such that a family $r$57 of $r$58-element sets contains three mutually disjoint sets whenever it satisfies

$r$59

and therefore contains a 3-sunflower whenever

$r$60

The proof uses extension generators, weighted $r$61-conditions, and recursive splitting arguments (Fukuyama, 2018).

The random greedy viewpoint leads to the sunflower-free process. Starting from the empty family and repeatedly adding a uniformly random $r$62-set that does not create an $r$63-sunflower, one obtains a random maximal sunflower-free family. When $r$64 with fixed $r$65, the process on an $r$66-element universe produces a $r$67-uniform $r$68-sunflower-free family of size

$r$69

where

$r$70

and $r$71 is the number of sunflowers containing a fixed $r$72-set. The analysis extends Bennett–Bohman’s theory of random greedy independent-set processes to hypergraphs with a sparse set of bad pairs of very large codegree (Bennett et al., 19 Sep 2025).

The sunflower property also appears in anti-Ramsey form. If the edges of a complete $r$73-uniform hypergraph are colored so that any monochromatic $r$74-sunflower has at most $r$75 petals, then sufficiently large vertex sets contain rainbow complete subhypergraphs; quantitatively, there exists a constant $r$76 depending only on $r$77 such that

$r$78

implies the existence of a rainbow subset of size at least $r$79 (Martínez-Sandoval et al., 2015).

Across these developments, a recurring theme is that the sunflower property sits at the boundary between local intersection data and global structure. In the set-theoretic setting, identical pairwise intersections already force the full sunflower geometry. In vector spaces, the failure of inclusion–exclusion for sums of three or more subspaces splits the theory into general-position and set-like variants. In random, tensor, and regularity-based methods, the core problem is to quantify when local sparsity or pseudorandomness forces the emergence of a common kernel.