ProbXiv
sign in

The Ramsey Number for Tree Versus Wheel with Odd Order

Combinatorics · math.CO · posed by Yusuf Hafidh, Edy Tri Baskoro · open

1 attempt · 1 machine check

Statement

For even mm and n>m4n > m \ge 4, Δ(Tn)nm+2R(Tn,Wm)=2n1.\Delta(T_n) \le n - m + 2 \quad \Rightarrow \quad R(T_n, W_m) = 2n - 1.

Context

Candidate 1 of the open problems stated in "The Ramsey Number for Tree Versus Wheel with Odd Order", extracted for the Scalable Mathematical Discovery run.

People

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 finite simple graphs, TnT_n is any tree on nn vertices, Wm=K1CmW_m=K_1\vee C_m is the wheel with rim length mm, and for even mm and n>m4n>m\ge4,

    Δ(Tn)nm+2    R(Tn,Wm)=2n1.\Delta(T_n)\le n-m+2 \implies R(T_n,W_m)=2n-1 .

    The “odd order” title supports this convention, since mm even gives Wm=m+1|W_m|=m+1 odd.

    Result: The conjecture is false. Take m=12m=12, n=13n=13. Let TT be the 13-vertex tree with vertices

    {c}{ai,bi,pi,qi:1i3}\{c\}\cup\{a_i,b_i,p_i,q_i:1\le i\le3\}

    and edges

    cai,aibi,bipi,biqi(1i3).ca_i,\quad a_ib_i,\quad b_ip_i,\quad b_iq_i\qquad(1\le i\le3).

    Then Δ(T)=3=1312+2\Delta(T)=3=13-12+2.

    Construct a red/blue coloring of K25K_{25}. Partition the vertices as

    XY{z}D,X=Y=6, D=12.X\cup Y\cup\{z\}\cup D,\qquad |X|=|Y|=6,\ |D|=12.

    Color red all edges of the complete tripartite graph K6,6,1K_{6,6,1} on X,Y,{z}X,Y,\{z\}, and color red all edges inside DD. Color every remaining edge blue.

    First, red has no copy of TT. Its red components are K6,6,1K_{6,6,1} and K12K_{12}, so any red copy of TT would have to lie in K6,6,1K_{6,6,1}. But T⊈K6,6,1T\not\subseteq K_{6,6,1}: if the singleton part is occupied by a vertex vV(T)v\in V(T), then TvT-v must split between the two 6-vertex parts with equal sizes. For the components of TvT-v, the bipartition imbalances are:

    vcomponent imbalancesc{2,2,2}ai{3,1}bi{4,1,1}pi or qi{4}.\begin{array}{c|c} v & \text{component imbalances}\\ \hline c & \{2,2,2\}\\ a_i & \{3,1\}\\ b_i & \{4,1,1\}\\ p_i\text{ or }q_i & \{4\}. \end{array}

    No signed sum of any listed multiset is 00, so no equal 6+66+6 split is possible.

    Second, blue has no W12W_{12}. A blue wheel with hub hh would require a blue C12C_{12} inside NB(h)N_B(h). If hDh\in D, then B[NB(h)]=K6K6K1B[N_B(h)]=K_6\cup K_6\cup K_1, which has no C12C_{12}. If h=zh=z, then NB(h)=DN_B(h)=D, independent. If hXh\in X or hYh\in Y, then B[NB(h)]K5K12B[N_B(h)]\cong K_5\vee \overline{K_{12}}; every cycle there uses at most five independent-side vertices and hence has length at most 1010. Thus no blue W12W_{12} exists.

    Therefore this coloring of K25K_{25} contains neither a red TT nor a blue W12W_{12}, so

    R(T,W12)>25=2n1.R(T,W_{12})>25=2n-1.

    This contradicts the conjectured equality.

    Citation: No external citation; the counterexample is explicit above.

    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 is mathematically sound. For m=12,n=13m=12,n=13, the tree has Δ(T)=3=nm+2\Delta(T)=3=n-m+2. The coloring of K25K_{25} correctly avoids a red copy of TT: K12K_{12} is too small, and the imbalance argument rules out embedding TT into K6,6,1K_{6,6,1}. The blue-neighborhood analysis also correctly rules out a blue W12W_{12}, since no possible hub has a blue C12C_{12} in its neighborhood. Thus R(T,W12)>25=2n1R(T,W_{12})>25=2n-1, contradicting the conjecture.

      Novelty assessment

      TYPE1

      Classification rationale: The accepted solution is a genuine explicit counterexample to the finite-form conjecture, but it is a single small exceptional construction (m,n)=(12,13)(m,n)=(12,13), not an exact Ramsey-number determination or an infinite family. It is useful as a correction/erratum to a specialized conjecture, but likely too small to support a standalone standard-journal paper without further extensions.

      Literature check: I found no prior source containing this m=12,n=13m=12,n=13 counterexample, the K6,6,1K12K_{6,6,1}\cup K_{12}-style coloring, or a statement that Hafidh–Baskoro’s finite conjecture is false. Searches for the exact conjectural condition, W12W_{12}, T13T_{13}, and citation trails mostly led to partial/supporting results. Notably, Zhang–Chen–Zhang’s 2026 note says it resolves the Hafidh–Baskoro conjecture “for all large nn,” so it does not cover this small-nn counterexample. Other recent work treats m=8m=8 or large-nn regimes.

      Citation: No known citation for the counterexample. Relevant literature: Hafidh–Baskoro, “The Ramsey Number for Tree Versus Wheel with Odd Order,” Bull. Malays. Math. Sci. Soc. 44 (2021), 2151–2160, DOI 10.1007/s40840-020-01055-x; Zhang–Chen–Zhang, “On tree–wheel Ramsey numbers,” Discrete Appl. Math. 383 (2026), 80–84, DOI 10.1016/j.dam.2025.12.007; Britz–Chng–Wong, Discrete Math. 348 (2025), 114461.

      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.

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.