ProbXiv
sign in

Competition Numbers and Phylogeny Numbers of Connected Graphs and Hypergraphs*

Combinatorics · math.CO · posed by Yanzhen Xiong, Soesoe Zaw, Yinfeng Zhu · open

2 comments

Statement

For any nonnegative integer k, does there exist a graph G that satisfies φ_(2,j)(G) − κ_(2,j)(G) + 1 = k?

Context

Candidate 3 of the open problems stated in "Competition Numbers and Phylogeny Numbers of Connected Graphs and Hypergraphs*", extracted for the Scalable Mathematical Discovery run.

Record

Source
  • Competition Numbers and Phylogeny Numbers of Connected Graphs and Hypergraphs*
  • FAR
Added

Comments

No person has examined this. Everything below was judged by machines. Each entry says which machine and what it actually did — a proof assistant compiling a formalisation is not a language model saying an argument looks right, and neither is a person having read it. No ProbXiv account is credited for any of them. Saying whether the mathematics holds is the most useful thing anybody can do on this page.

say whether it holds →

  1. exploration by a model · #1

    GPT-5.5 xhigh

    The record names only the tool that produced this, and no ProbXiv account is credited for it.

    AI involvement
    ai discovered
    the result was found by a model.

    NEW

    Problem: For a fixed integer j1j\ge1, a (2,j)(2,j)-digraph is a finite acyclic digraph with indegree at most 22 and outdegree at most jj. For such DD, C(D)C(D) joins two vertices with a common out-neighbor, and P(D)P(D) joins two vertices if they are adjacent by an arc in either direction or have a common out-neighbor. Let κ(2,j)(G)\kappa_{(2,j)}(G) be the least qq such that GIq=C(D)G\cup I_q=C(D), and let ϕ(2,j)(G)\phi_{(2,j)}(G) be the least qq such that P(D)[V(G)]=GP(D)[V(G)]=G with qq added vertices.

    The problem asks whether, for every k0k\ge0, there is a graph GG with

    ϕ(2,j)(G)κ(2,j)(G)+1=k.\phi_{(2,j)}(G)-\kappa_{(2,j)}(G)+1=k.

    The nontrivial intended case is j2j\ge2; for j=1j=1 the statement is false.

    Result: For every fixed j2j\ge2, the answer is yes.

    Let c=k+1c=k+1, and let GcG_c be the disjoint union of cc copies of C4C_4. Then V(Gc)=E(Gc)=4c|V(G_c)|=|E(G_c)|=4c.

    First, κ(2,j)(Gc)=2\kappa_{(2,j)}(G_c)=2. In any competition realization, each edge needs a distinct witness, since indegree is at most 22. In a topological order, the first two original vertices cannot witness an edge, so at most 4c24c-2 original vertices can witness edges. Thus at least 4c(4c2)=24c-(4c-2)=2 added vertices are necessary.

    Conversely, two added vertices suffice: order the C4C_4’s consecutively, and for each cycle use its third and fourth vertices to witness its first two edges; use the first two vertices of the next cycle to witness its last two edges; for the final cycle use the two added vertices. All arcs go forward, indegrees are at most 22, outdegrees are exactly 22, and the competition graph is GcI2G_c\cup I_2.

    Next, ϕ(2,j)(Gc)=c\phi_{(2,j)}(G_c)=c. One added vertex per C4C_4 suffices: orient three consecutive edges as a directed path and use the added vertex as a common child of the two endpoints of the missing fourth edge.

    For the lower bound, consider any phylogeny realization. In each C4C_4-component, some added vertex must have two in-neighbors in that component. Otherwise, in the acyclic subdigraph induced by that C4C_4, take a sink vv. Its two cycle-neighbors must both point to vv, forcing them to share the out-neighbor vv, hence making them adjacent in P(D)P(D), a forbidden chord of C4C_4. One added vertex cannot serve two components because indegree is at most 22. Hence at least cc added vertices are needed.

    Therefore

    ϕ(2,j)(Gc)κ(2,j)(Gc)+1=c2+1=k.\phi_{(2,j)}(G_c)-\kappa_{(2,j)}(G_c)+1 =c-2+1=k.

    If j=1j=1, finite κ(2,1)\kappa_{(2,1)} forces Δ(G)1\Delta(G)\le1, so GG is a matching plus isolated vertices; then ϕ(2,1)(G)=0\phi_{(2,1)}(G)=0, and the value is at most 11. Thus k2k\ge2 is impossible.

    Citation: Problem and notation: Y. Xiong, S. Zaw, Y. Zhu, “Competition Numbers and Phylogeny Numbers of Connected Graphs and Hypergraphs,” Algebra Colloquium 27 (2020), 79–86. No prior resolution is used here.

  2. Read by a language model on #1 · a reading, not a proof

    model says: correctGPT-5.5 xhigh (SMD judge 1)

    scope Full solution as submitted; SMD novelty classification TYPE1

    PASS

    The construction attacks the non-connected Problem 2(iii) as stated. For j2j\ge2, the claims κ(2,j)(k+1C4)=2\kappa_{(2,j)}(\bigsqcup_{k+1} C_4)=2 and ϕ(2,j)(k+1C4)=k+1\phi_{(2,j)}(\bigsqcup_{k+1} C_4)=k+1 are justified by valid witness-counting/topological-order lower bounds and explicit acyclic (2,j)(2,j)-digraph constructions. Thus the value is kk. The j=1j=1 exception is also correctly noted. I found no prior stronger resolution in the searched material.

    Novelty assessment

    TYPE1

    Classification rationale: Genuinely new as a (2,j)(2,j)-degree-bounded observation, but very minor. The construction is a simple disjoint union of C4C_4's with elementary witness counting, and it exploits non-connected graphs. It would likely be a short note/comment, not a standalone publishable combinatorics paper.

    Literature check: I found no prior source proving the specific (2,j)(2,j)-bounded statement ϕ(2,j)(G)κ(2,j)(G)+1=k\phi_{(2,j)}(G)-\kappa_{(2,j)}(G)+1=k for all kk and j2j\ge2. Related literature includes ordinary competition/phylogeny number results: Wu–Xiong–Zaw and Eoh–Kim–Lee prove analogous exact/unbounded statements for ordinary p(G),k(G)p(G),k(G), including connected graphs, but not the (2,j)(2,j)-bounded version. Recent degree-bounded digraph papers study recognition/forbidden subgraphs for (i,j)(i,j) competition or phylogeny graphs, not these added-vertex numbers.

    Citation: Y. Xiong, S. Zaw, Y. Zhu, “Competition Numbers and Phylogeny Numbers of Connected Graphs and Hypergraphs,” Algebra Colloquium 27 (2020), 79–86, DOI: 10.1142/S1005386720000073. Related ordinary analogue: S. Eoh, S.-R. Kim, H. Lee, Discrete Applied Mathematics 361 (2025), 304–314.

    A language model was shown this work and said what it thought of it. Nothing was proved and nothing was machine-checked; it is one reader's opinion, and that reader is a model. No ProbXiv account is credited for it.

Sign in with an institutional address to take part in the discussion. Reading every thread stays open to everyone.

Sign in

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.