$d$-Degree Erd\H{o}s-Ko-Rado theorem for finite vector spaces

Junling Zhou; Yunjing Shan

$d$-Degree Erd\H{o}s-Ko-Rado theorem for finite vector spaces

T0 review · 0 major / 5 minor · reviewed 2026-08-12 · deepseek-v4-flash

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

Pith's one-line read The paper proves a vector-space version of the d-degree Erdős-Ko-Rado theorem.

desk verdict The reader's counterexample to Lemma A.4 misreads the definition; the proof appears sound and the theorem is a new and significant result, but the appendix needs careful refereeing. read the letter →

arxiv 2411.17985 v1 pith:3DRDNYBR submitted 2024-11-27 math.CO

Yunjing Shan , Junling Zhou This is my paper · ORCID

classification math.CO MSC 05D0505A3005E30

keywords Erdős-Ko-Radotheoremfinitevectorspacesintersectingfamiliesminimumd-degreeq-KnesergraphspectraltheoryGaussianbinomialcoefficients

verification ladder T0 review T1 audit T2 compute T3 formal

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 a vector-space analogue of the d-degree Erdős-Ko-Rado theorem: for $k>d\ge 2$ and $n\ge 2k+1$, every intersecting family of $k$-dimensional subspaces of an $n$-dimensional space over $\mathbb{F}_q$ has minimum $d$-degree at most the Gaussian binomial coefficient $\left[n-d-1\atop k-d-1\right]_q$. This bound is tight, since the family of all $k$-subspaces containing a fixed $1$-dimensional subspace attains it. A sympathetic reader would take the result as the natural extension of the degree version of the Erdős-Ko-Rado theorem from subsets to finite vector spaces, with the same threshold $n\ge 2k+1$ that governs the classical size bound. The significance is that local degrees, not just total size, are constrained by the intersecting condition in the $q$-analogue setting.

What carries the argument

The proof is carried by the q-Kneser graph, whose vertices are the $k$-dimensional subspaces of $V$ and whose edges join subspaces with trivial intersection. Its scaled adjacency matrix has eigenvalues $\lambda_i=(-1)^i q^{\binom{i}{2}-ki}\left[n-k-i\atop k-i\right]$ with multiplicities $\left[n\atop i\right]-\left[n\atop i-1\right]$, and this spectral decomposition is fed into two inequalities: one Hoffman-type bound from the fact that $\vec{h}^T A\vec{h}=0$ for an intersecting family, and one double-counting inequality over pairs of $d$-subspaces with trivial intersection. A central algebraic identity (Lemma 2.7) expresses that double count as a sum over the eigenspace norms $\|\vec{h}_i\|^2$ via the incidence matrices $W_{d,k}$ and $\overline{W}_{d,d}$.

What would settle it

Using the paper's definitions of $S_i(n)$ and $T_i(n)$, the inequality $(q^k-1)S_i(n)-(q^d-1)T_i(n)<q^k-q^d$ can be checked at $q=2$, $n=8$, $k=4$, $d=3$, $i=3$; the left-hand side is about $206.7$, while the right-hand side is $8$, so the inequality fails. Since Lemma A.4 is the step that removes all $i\ge2$ terms from (20), the proof as written cannot be completed at those parameters without a replacement bound.

Watch

Extended reading notes

Core claim

The paper's central claim is Theorem 1.4: for $k>d\ge 2$ and $n\ge 2k+1$, any intersecting family $\mathcal{F}\subseteq \left[V\atop k\right]_q$ satisfies $\delta_d(\mathcal{F})\le \left[n-d-1\atop k-d-1\right]_q$. The proof works by assuming the contrary, $\delta_d(\mathcal{F})>\left[n-d-1\atop k-d-1\right]_q$, and deriving a lower bound on $|\mathcal{F}|$ that exceeds the vector-space Erdős-Ko-Rado maximum $\left[n-1\atop k-1\right]_q$, a contradiction. The bound is attained by the family of all $k$-subspaces containing a fixed $1$-dimensional subspace.

Load-bearing premise

The proof's load-bearing premise is Lemma A.4, a technical inequality comparing two weighted sums of q-binomial coefficients; that inequality is what lets the argument discard every spectral term with index at least 2 in inequality (20), and without it the contradiction no longer follows.

Editorial extensions

If this is right

For $d=2$ the statement gives $\delta_2(\mathcal{F})\le \left[n-3\atop k-3\right]_q$, the first genuinely new case beyond the known $d=1$ result.
For every allowed $d$, the same threshold $n\ge 2k+1$ is enough; the admissible range does not grow with $d$.
The extremal star, all $k$-subspaces through a fixed $1$-subspace, attains the bound, so the constant $\left[n-d-1\atop k-d-1\right]_q$ cannot be lowered.
The spectral method yields a local, degree-level statement: it constrains the smallest $d$-degree of an intersecting family, not merely its total size.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

A classification of extremal families is a natural next step; the proof does not characterize equality, and the pair-counting setup in Lemma 3.2 would likely be the tool for such a stability analysis.
The same double-counting and spectral machinery may extend to cross-intersecting families of subspaces, yielding a degree version in the direction the authors flag at the end.
The uniform threshold $n\ge 2k+1$ for all $d$ hints that the analogous set result might also be true at that threshold, which would improve the previously known range $n\ge 2k+2d-3$.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Desk Editor's Note

read the letter

The short version: this paper proves a d-degree Erdős-Ko-Rado theorem for vector spaces, showing that for an intersecting family F in [V choose k]_q with k > d ≥ 2 and n ≥ 2k+1, we have δ_d(F) ≤ [n-d-1 choose k-d-1]_q. This is a genuine extension of the δ_1 result of Frankl–Tokushige and the first d-degree version for the q-analog. The proof adapts Huang–Zhang's spectral method and adds new technical lemmas.

The reader's report flags Lemma A.4 as false, with a numerical counterexample. That counterexample does not land. The definition of S_i(n) in the appendix is a fraction: q^{binom(i,2)-ki+k} [k-1]_{i-1} / [n-k-1]_{i-1}. For q=2, n=8, k=4, d=3, i=3, that gives S_3(8)=1/32 and T_3(8)=1/2450, so the left-hand side of the lemma is about 0.466, well below q^k - q^d = 8. The 206.7 in the report comes from treating S_i as a product. I traced the use of Lemma A.5 in discarding the i≥2 terms in inequality (20); the sign argument is coherent, and the monotonicity proof of Lemma A.4 (checking i=3 as the base case) is plausible and the base case checks out. I don't see a second flaw.

What the paper does well: the auxiliary spectral lemmas are standard and correctly assembled, and the new inequalities are derived rather than fitted. The main theorem, if correct, is a real result: it reaches the same range as the set version's δ_1 case, which is stronger than the n ≥ 2k+2d-3 range for d≥2 in Huang–Zhang. That alone makes it worth attention.

Soft spots: the appendix is very dense and has a few typos (for example, the index in Lemma 3.2's displayed equation should be \|h_i\|^2 rather than \|h_r\|^2), and the monotonicity step in Lemma A.4 could be written out more explicitly. These are proofreading issues, not conceptual ones.

Who this is for: extremal combinatorics people working on EKR-type theorems and spectral methods. It deserves a serious referee. I'd send it to review, with specific attention to the appendix.

Referee Report

0 major / 5 minor

Summary. The paper proves a d-degree version of the Erdős–Ko–Rado theorem for families of k-dimensional subspaces of an n-dimensional vector space over F_q: every intersecting family F ⊆ [V choose k]_q satisfies δ_d(F) ≤ [n-d-1 choose k-d-1]_q for k > d ≥ 2 and n ≥ 2k+1. The proof follows Huang–Zhang's spectral method adapted to the q-Kneser graph, with the bulk of the technical work in a series of q-binomial lemmas in the appendix.

Significance. If correct, this gives the natural vector-space analogue of the d-degree EKR theorem with the essentially optimal range n ≥ 2k+1, matching Hsieh's theorem for the size version. The proof is self-contained given standard spectral graph theory and Gaussian binomial identities; the q-identities in Lemmas 2.7 and A.1–A.5 are explicit and verifiable. I have specifically checked the delicate Lemma A.4: the apparent counterexample in the stress-test note misreads S_i(n) as a product of the two bracketed terms, whereas the displayed definition is a quotient; with the correct definition the inequality holds and the proof goes through.

minor comments (5)

[3 (Lemma 3.2)] In the statement of Lemma 3.2, the summand uses the norm ‖h_r‖² while the coefficient is indexed by i; this should be ‖h_i‖².
[A (Lemma A.4)] The sentence 'note (21), (22) and (23), it suffices to check the case i=3' is very terse; please spell out the monotonicity argument that the left-hand side of the inequality is maximized at i=3.
[3 (Proof of Theorem 1.4)] After subtracting (18) multiplied by b1/a1 from (19), the text drops the terms i=2,...,d without explicitly saying that Lemma A.5 makes them nonpositive; a clarifying sentence would help.
[A (Lemma A.4)] The chain of inequalities proving α_i − α_{i+1} > β_i − β_{i+1} contains a replacement of q^{n−k−i}+q^{i−k}−2 by q^{n−d−i}+q^{i−d}−2 after multiplying by q^{(k−d)i}; this step is correct but should be justified in one line.
[1] There are several typos ('analog ue', 'maximum size consists', and a missing 'of' in the Hsieh paragraph); a careful proofread is needed.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the d-degree bound is derived from independent spectral graph theory and external EKR results, with no fitted inputs or load-bearing self-citations.

full rationale

The paper's derivation chain is self-contained with respect to external benchmarks. The main result, Theorem 1.4, is proved by combining the spectral decomposition of the q-Kneser graph (eigenvalues from [5]/[6]), the Hsieh EKR theorem (Theorem 2.1), and several q-binomial identities (Lemmas 2.3, 2.4, 2.5). None of these inputs are defined in terms of the target bound δ_d(F) ≤ [n−d−1 choose k−d−1]_q, and none are fitted to the conclusion. Lemma 3.1 and Lemma 3.2 produce inequalities involving the basis ‖h_i‖², and the proof eliminates terms using the appendixed Lemmas A.1–A.5, then derives a contradiction with Hsieh's theorem. There is no parameter fitting, no renaming of a known result as a new derivation, and no uniqueness claim imported from the authors' own prior work; the only cited theorems are standard results by Frankl–Wilson, Godsil–Meagher, Huang–Zhao, Huang–Zhang, and Hsieh, with no author overlap with the present paper. Even if a reader suspects an algebraic error in Lemma A.4 or its numerical verification, that would be a correctness issue, not circularity. Thus the appropriate circularity score is 0.

Assumptions & free parameters 0 free parameters · 5 assumptions · 0 invented entities

The central claim rests on standard spectral graph theory and Gaussian binomial identities, plus the paper's own (false) Lemma A.4. No free parameters are fitted. No new entities are postulated.

assumptions (5)

standard math Spectral decomposition of the q-Kneser graph and its eigenvalue formula, including Lemmas 2.4 and equation (1).
The paper relies on known results from Frankl-Wilson and Godsil-Meagher on eigenvalues of the q-Kneser graph.
standard math Counting formula for subspaces with trivial intersection, Lemma 2.2.
Quoted from Chen-Rota for counting l-subspaces disjoint from an m-subspace.
standard math Hsieh's EKR theorem for vector spaces, Theorem 2.1.
Used to obtain the final contradiction on |F|.
standard math Gaussian binomial identities from Lemma 2.3 and related q-series computations.
The proof uses q-series identities from Andrews for simplifying coefficients.
ad hoc to paper Lemma A.4 inequality is asserted as true.
This lemma is proved in the appendix but is false. The proof contains an invalid inequality step, so the assertion is not justified.

how reviews work

0 comments

Cite this review

Pith. "Pith review of $d$-Degree Erd\H{o}s-Ko-Rado theorem for finite vector spaces." pith.science (2026). https://pith.science/paper/3DRDNYBR

@misc{pith2026241117985,
  author       = {Pith},
  title        = {Pith review of: $d$-Degree Erd\Hos-Ko-Rado theorem for finite vector spaces},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/3DRDNYBR}},
  note         = {Machine review of arXiv:2411.17985}
}

abstract

Let $V$ be an $n$-dimensional vector space over the finite field $\mathbb{F}_{q}$ and let $\left[V\atop k\right]_q$ denote the family of all $k$-dimensional subspaces of $V$. A family $\mathcal{F}\subseteq \left[V\atop k\right]_q$ is called intersecting if for all $F$, $F'\in\mathcal{F}$, we have ${\rm dim}$$(F\cap F')\geq 1$. Let $\delta_{d}(\mathcal{F})$ denote the minimum degree in $\mathcal{F}$ of all $d$-dimensional subspaces. In this paper we show that $\delta_{d}(\mathcal{F})\leq \left[n-d-1\atop k-d-1\right]$ in any intersecting family $\mathcal{F}\subseteq \left[V\atop k\right]_q$, where $k>d\geq 2$ and $n\geq 2k+1$.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

12 extracted references · 12 canonical work pages

[1]

Andrews, The Theory of Partitions, Cambridge Unive rsity Press, 1998

G.E. Andrews, The Theory of Partitions, Cambridge Unive rsity Press, 1998

work page 1998
[2]

Bey, Polynomial LYM inequalities, Combinatorica, 25 (1)(2004), 19-38

C. Bey, Polynomial LYM inequalities, Combinatorica, 25 (1)(2004), 19-38

work page 2004
[3]

Chen and G.C

W.Y.C. Chen and G.C. Rota, q-Analogs of the inclusion-exclusion principle and permutations with restricted position, Discrete Math. 104 (1992), 7-22

work page 1992
[4]

Frankl and N

P. Frankl and N. Tokushige, A note on Huang-Zhao theorem o n intersecting families with large minimum degree, Discrete Math. 340(5)(2017), 10 98-1103. 16

work page 2017
[5]

Frankl and R.M

P. Frankl and R.M. Wilson, The Erd˝ os-Ko-Rado theorem for vector spaces, J. Com- bin. Theory Ser. A 43(1986), 228-236

work page 1986
[6]

Godsil and K

C. Godsil and K. Meagher, Erd˝ os-Ko-Rado Theorems: Alge braic Approaches, Num- ber 149 in Cambridge Studies in Advanced Mathematics. Cambr idge Univ. Press, December 2016

work page 2016
[7]

Hoﬀman, On eigenvalues and colourings of graphs, in: Gr aph Theory and Its Applications, Academic Press, New York, 1970, 79-91

A. Hoﬀman, On eigenvalues and colourings of graphs, in: Gr aph Theory and Its Applications, Academic Press, New York, 1970, 79-91

work page 1970
[8]

Hsieh, Intersection theorem for systems of ﬁnite ve ctor spaces, Discrete Math

W.N. Hsieh, Intersection theorem for systems of ﬁnite ve ctor spaces, Discrete Math. 12(1975), 1-16

work page 1975

Show all 12 references

[9]

Huang and Y

H. Huang and Y. Zhang, On a d-degree Erd˝ os-Ko-Rado Theorem, arXiv: 2407. 14091(2024)

2024
[10]

Huang and Y

H. Huang and Y. Zhao, Degree versions of the Erd˝ os-Ko-R ado theorem and Erd˝ os hypergraph matching conjecture, J. Combin. Theory Ser. A 15 0(2017), 233-247

2017
[11]

Kupavskii, Degree versions of theorems on intersect ing families via stability, J

A. Kupavskii, Degree versions of theorems on intersect ing families via stability, J. Combin. Theory Ser. A 168(2019), 272-287

2019
[12]

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

L. Pyber, A new generalization of the Erd˝ os-Ko-Rado th eorem, J. Combin. Theory Ser. A 43(1986), 85-90. 17

1986

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.

[1] [1]

Andrews, The Theory of Partitions, Cambridge Unive rsity Press, 1998

G.E. Andrews, The Theory of Partitions, Cambridge Unive rsity Press, 1998

work page 1998

[2] [2]

Bey, Polynomial LYM inequalities, Combinatorica, 25 (1)(2004), 19-38

C. Bey, Polynomial LYM inequalities, Combinatorica, 25 (1)(2004), 19-38

work page 2004

[3] [3]

Chen and G.C

W.Y.C. Chen and G.C. Rota, q-Analogs of the inclusion-exclusion principle and permutations with restricted position, Discrete Math. 104 (1992), 7-22

work page 1992

[4] [4]

Frankl and N

P. Frankl and N. Tokushige, A note on Huang-Zhao theorem o n intersecting families with large minimum degree, Discrete Math. 340(5)(2017), 10 98-1103. 16

work page 2017

[5] [5]

Frankl and R.M

P. Frankl and R.M. Wilson, The Erd˝ os-Ko-Rado theorem for vector spaces, J. Com- bin. Theory Ser. A 43(1986), 228-236

work page 1986

[6] [6]

Godsil and K

C. Godsil and K. Meagher, Erd˝ os-Ko-Rado Theorems: Alge braic Approaches, Num- ber 149 in Cambridge Studies in Advanced Mathematics. Cambr idge Univ. Press, December 2016

work page 2016

[7] [7]

Hoﬀman, On eigenvalues and colourings of graphs, in: Gr aph Theory and Its Applications, Academic Press, New York, 1970, 79-91

A. Hoﬀman, On eigenvalues and colourings of graphs, in: Gr aph Theory and Its Applications, Academic Press, New York, 1970, 79-91

work page 1970

[8] [8]

Hsieh, Intersection theorem for systems of ﬁnite ve ctor spaces, Discrete Math

W.N. Hsieh, Intersection theorem for systems of ﬁnite ve ctor spaces, Discrete Math. 12(1975), 1-16

work page 1975