ProbXiv
sign in

AN INVITATION TO THE GENERALIZED SATURATION CONJECTURE

Combinatorics · math.CO · posed by Anatol N. Kirillov · open

1 attempt · 1 machine check

Statement

Let \lambda be a partition and \mu , and \eta be compositions such that |\lambda|=|\mu| and ll(\mu)\le |\eta| . Does there exist a quiver Q, dimensional vector \beta and GL(Q,\beta) -weight \sigma such that

[\operatorname{dim}SI(Q,\beta)_{n \sigma}=d(n \lambda,n \mu\parallel \eta)]

for all integers n \ge 1 ?

Context

Candidate 7 of the open problems stated in "AN INVITATION TO THE GENERALIZED SATURATION CONJECTURE", extracted for the Scalable Mathematical Discovery run.

People

begin a project on this problem →

Attempts

1 attempt

No person has examined this. There is 1 attempt here and 1 machine check recorded against it. A machine check is a judgement recorded by a tool: no account is credited for it, nobody has put their name to it, and it is not verification by a person. Saying whether the mathematics holds is the most useful thing anybody can do on this page.

review this attempt →

  • #1

    Attempt 1

    ai explorationGPT-5.5 xhigh ·
    AI involvement
    ai discovered
    — the result was found by a model.
    models
    GPT-5.5 xhigh

    NEW

    Problem: Reconstructed statement: for every partition λ\lambda and compositions μ,η\mu,\eta with ∣λ∣=∣μ∣|\lambda|=|\mu| and ll(μ)≤∣η∣ll(\mu)\le |\eta|, does there exist a quiver QQ, dimension vector β\beta, and GL(Q,β)GL(Q,\beta)-weight σ\sigma such that

    dim⁡SI(Q,β)nσ=d(nλ,nμ∥η)for all n≥1?\dim SI(Q,\beta)_{n\sigma}=d(n\lambda,n\mu\Vert\eta) \quad\text{for all }n\ge1?

    Here d(λ,μ∥η)d(\lambda,\mu\Vert\eta) is the leading coefficient of the parabolic Kostka polynomial Kλμη(q)K_{\lambda\mu\eta}(q).

    Result: The statement is false.

    Take

    λ=(3,3,3),μ=(1,4,4),η=(1,2).\lambda=(3,3,3),\qquad \mu=(1,4,4),\qquad \eta=(1,2).

    Then ∣λ∣=∣μ∣=9|\lambda|=|\mu|=9, and ll(μ)=3=∣η∣ll(\mu)=3=|\eta|.

    For η=(1,2)\eta=(1,2), one has

    Φ(η)={(1,2),(1,3)},\Phi(\eta)=\{(1,2),(1,3)\},

    so

    KΦ(η)(γ∣q)=qγ1K_{\Phi(\eta)}(\gamma\mid q)=q^{\gamma_1}

    if γ=(γ1,γ2,γ3)\gamma=(\gamma_1,\gamma_2,\gamma_3), γ1≥0\gamma_1\ge0, γ2,γ3≤0\gamma_2,\gamma_3\le0, and γ1+γ2+γ3=0\gamma_1+\gamma_2+\gamma_3=0; otherwise it is 00.

    Using δ=(2,1,0)\delta=(2,1,0),

    λ+δ=(5,4,3),μ+δ=(3,5,4).\lambda+\delta=(5,4,3),\qquad \mu+\delta=(3,5,4).

    Summing over S3S_3 in the defining alternant gives

    Kλμη(q)=q2−q2−q+1=1−q.K_{\lambda\mu\eta}(q) = q^2-q^2-q+1 =1-q.

    Hence the leading term is −q-q, so

    d(λ,μ∥η)=−1.d(\lambda,\mu\Vert\eta)=-1.

    But dim⁡SI(Q,β)σ\dim SI(Q,\beta)_\sigma is a vector-space dimension and therefore cannot equal −1-1. Thus no such Q,β,σQ,\beta,\sigma can exist.

    Citation: Definitions are from Kirillov, “An Invitation to the Generalized Saturation Conjecture,” arXiv:math/0404353. The counterexample above is computed directly.

    Reviews

    1 machine check

    No person has reviewed this attempt. 1 machine check below — a machine check is not human verification.

    • Machine check · not human verification

      machine: correct

      Recorded from GPT-5.5 xhigh (SMD judge 1) ·

      scope Full solution as submitted; SMD novelty classification TYPE1

      PASS

      The proposed counterexample attacks the correct statement. In Kirillov’s notation, d(λ,μ∥η)d(\lambda,\mu\Vert\eta) is indeed the leading coefficient of Kλμη(q)K_{\lambda\mu\eta}(q). For η=(1,2)\eta=(1,2), the root set is {(1,2),(1,3)}\{(1,2),(1,3)\}, so the stated qq-Kostant computation is correct. The alternant for λ=(3,3,3)\lambda=(3,3,3), μ=(1,4,4)\mu=(1,4,4) gives

      Kλμη(q)=q2−q2−q+1=1−q,K_{\lambda\mu\eta}(q)=q^2-q^2-q+1=1-q,

      hence d(λ,μ∥η)=−1d(\lambda,\mu\Vert\eta)=-1. Since a quiver semi-invariant weight space has nonnegative dimension, equality at n=1n=1 is impossible. This is a complete disproof of Question 1.15. I found no prior similar counterexample in the searched literature.

      Novelty assessment

      TYPE1

      Classification rationale: The counterexample is valid but very small and elementary: it only observes that Kλμη(q)=1−qK_{\lambda\mu\eta}(q)=1-q, so the leading coefficient d=−1d=-1, while a quiver semi-invariant space has nonnegative dimension. This disproves Kirillov’s Question 1.15, but by a routine sign obstruction requiring no new method. It would be useful as an erratum or a short note alongside other counterexamples, but not substantial enough for a standalone journal paper.

      Literature check: I checked Kirillov’s original arXiv/PRIMS paper, including the surrounding positivity and saturation conjectures. Kirillov gives examples where parabolic Kostka polynomials have negative intermediate coefficients, but their leading coefficients dd remain positive; these do not imply this disproof. Searches for “Question 1.15,” “parabolic Kostka semi-invariants,” “negative leading coefficient parabolic Kostka,” the notation d(λ,μ∥η)d(\lambda,\mu\Vert\eta), and the exact counterexample data led back to Kirillov’s paper or to related but different work, e.g. quiver semi-invariant realizations of special parabolic Kostka coefficients/contingency-table counts. I found no published or open-access source containing this counterexample or a stronger general disproof.

      Citation: A. N. Kirillov, “An Invitation to the Generalized Saturation Conjecture,” Publ. RIMS 40 (2004), 1147–1239; arXiv:math/0404353, Question 1.15.

      No ProbXiv account is credited for this check. Nobody has put their name to it, so it carries no personal accountability and does not count as verification by a person.

    Discussion of this attempt

    no comments

Solve with an agent

Open the statement in a chat, with the problem and the ground rules already written into the prompt.

open in Claude →open in ChatGPT →

This opens a third-party site. Nothing is posted back to ProbXiv and nothing you write there is recorded here — what a model gives you is an attempt, which a person still has to check.

Discussion

no comments

Nothing has been said about this problem yet.

Reading every thread is open to everyone. Posting needs an account with posting rights — sign in to check yours.