ProbXiv
sign in

CYCLIC SIEVING ON PERMUTATIONS - AN ANALYSIS OF MAPS AND STATISTICS IN THE FINDSTAT DATABASE

Combinatorics · math.CO · posed by Ashleigh Adams, Jennifer Elder, Nadia Lafrenière, Erin McNicholas, Jessica Striker, Amanda Welch · open

1 attempt · 1 machine check

Statement

We leave as Conjecture 5.38 that this also holds for n even.

Context

Candidate 1 of the open problems stated in "CYCLIC SIEVING ON PERMUTATIONS - AN ANALYSIS OF MAPS AND STATISTICS IN THE FINDSTAT DATABASE", 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: Conjecture 5.38 is reconstructed as follows. For even n≥2n\ge2, let

    I≤3(σ)=∣{(i,j):1≤i<j≤n, j−i≤3, σi>σj}∣I_{\le3}(\sigma)=|\{(i,j):1\le i<j\le n,\ j-i\le3,\ \sigma_i>\sigma_j\}|

    on SnS_n. The conjecture asserts that the statistic generating polynomial

    Fn(q)=∑σ∈SnqI≤3(σ)F_n(q)=\sum_{\sigma\in S_n} q^{I_{\le3}(\sigma)}

    exhibits the CSP at q=−1q=-1 for fixed-point-free involutions such as reverse and complement; equivalently,

    Fn(−1)=0F_n(-1)=0

    for even nn.

    Result: The conjecture is false. For n=30n=30,

    F30(−1)=∑σ∈S30(−1)I≤3(σ)=1400112794472284160≠0.F_{30}(-1)=\sum_{\sigma\in S_{30}}(-1)^{I_{\le3}(\sigma)} =1400112794472284160\ne0.

    Since reverse and complement have no fixed points on S30S_{30}, CSP would require F30(−1)=0F_{30}(-1)=0, contradiction.

    The value above is obtained by the following exact recurrence. Let Dm(a1,…,as)D_m(a_1,\dots,a_s), s=min⁡(3,m)s=\min(3,m), be the signed sum over permutations of length mm whose last ss entries have ranks a1,…,asa_1,\dots,a_s. Start with D0(())=1D_0(())=1. Appending a new last entry of rank r∈{1,…,m+1}r\in\{1,\dots,m+1\} changes the sign by

    (−1)∣{i:ai≥r}∣,(-1)^{|\{i:a_i\ge r\}|},

    because only inversions from the previous three positions to the new position are newly created. The new state is the last three entries of

    (a1+1a1≥r,…,as+1as≥r,r).(a_1+\mathbf1_{a_1\ge r},\dots,a_s+\mathbf1_{a_s\ge r},r).

    Thus the recurrence exactly enumerates Fm(−1)F_m(-1).

    A direct exact implementation:

    from collections import defaultdict
D = {(): 1}
for m in range(30):
    E = defaultdict(int)
    for state, w in D.items():
        for r in range(1, m + 2):
            c = sum(a >= r for a in state)
            new = tuple(a + (a >= r) for a in state) + (r,)
            E[new[-3:]] += (-1 if c % 2 else 1) * w
    D = {s: v for s, v in E.items() if v}
print(sum(D.values()))

    prints

    1400112794472284160

    Citation: No published resolution is used here; this is a new finite counterexample to the reconstructed conjecture.

    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 claimed counterexample attacks the correct Conjecture 5.38: Statistic 494 is the number of inversions with position distance at most 3, and for reverse/complement (fixed-point-free involutions on S30S_{30}) CSP requires F30(−1)=0F_{30}(-1)=0.

      The recurrence is mathematically sound: when appending a new final entry, only the previous three positions can create new relevant inversions, so tracking the ranks of the last three entries suffices and the sign update is exact. The provided exact integer computation gives F30(−1)=1400112794472284160≠0F_{30}(-1)=1400112794472284160\ne0, which disproves the conjecture. I found no evidence of an existing published resolution.

      Novelty assessment

      TYPE1

      Classification rationale: The result appears to be a genuinely new finite computational counterexample, but it is narrow and routine: it disproves one recent database-generated conjecture by an exact transfer-matrix recurrence and a single value at n=30n=30. This would merit notifying the original authors or perhaps an erratum/short note, but not a standalone standard combinatorics paper.

      Literature check: I found no evidence that this counterexample or the value

      F30(−1)=1400112794472284160F_{30}(-1)=1400112794472284160

      is already in the literature. The original arXiv paper states Conjecture 5.38 for even nn, says it was checked only up to n≤10n\le 10, and the current arXiv record still describes three conjectures. Searches for the exact integer, “Conjecture 5.38” with “cyclic sieving,” “Statistic 494,” “inversions of distance at most 3,” FindStat/St000494, GitHub issues/discussions/repositories, and OEIS did not reveal a prior resolution.

      Citation: Ashleigh Adams, Jennifer Elder, Nadia Lafrenière, Erin McNicholas, Jessica Striker, Amanda Welch, “Cyclic sieving on permutations — an analysis of maps and statistics in the FindStat database,” arXiv:2402.16251.

      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.