Pith. sign in

REVIEW 1 major objections 2 minor 6 references

Order plus size of $\tau$-critical graphs

T0 review · 1 major / 2 minor · reviewed 2026-08-14 · deepseek-v4-flash

Pith's one-line read The paper proves that every τ-critical graph G of order n with τ(G)=t satisfies |V|+|E| ≤ {t+2 choose 2}, with equality exactly for K_n (n≥2), 2K2, and C5.

desk verdict The combined bound is clean and correct, but the equality-case proof contains a false claim about 2-regular complements that needs fixing before publication. read the letter →

arxiv 1908.05225 v1 pith:7ZP4EAA7 submitted 2019-08-13 math.CO

classification math.CO MSC 05C3505C69
keywords τ-criticalgraphstransversalnumbervertexcoverextremalgraphtheorycombinedorderandsizeboundHajnaldegreeequalitycases
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 answers a natural extremal question: for graphs that are minimally imperfect with respect to the vertex-cover number, how large can the sum of the number of vertices and edges be? It establishes the sharp bound |V|+|E| ≤ {t+2 choose 2} for every τ-critical graph with τ(G)=t, improving the separate bounds |V| ≤ 2t and |E| ≤ {t+1 choose 2} by combining them into a single inequality. A sympathetic reader should care because such combined bounds often expose structure that separate bounds hide; here the extremal graphs turn out to be exactly three simple families: complete graphs, a disjoint union of two edges, and a five-cycle.

What carries the argument

The load-bearing tool is Hajnal's degree bound (Theorem A): in a τ-critical graph of order n with τ(G)=t, every vertex x satisfies deg_G(x) ≤ 2t−n+1. This bound is what turns the global question about the number of edges into a local degree check. Summing it over all vertices yields the edge estimate that, after adding |V|=n, reduces the combined inequality to the quadratic (n−t−1)(n−t−2) ≥ 0. The equality analysis then uses the fact that equality in the bound forces n=t+1 or n=t+2, and in the latter case the complement must be 2-regular; requiring τ-criticality forces the complement to be a single cycle of length at most five, giving the three stated extremal graphs.

What would settle it

Find a τ-critical graph G with τ(G)=t such that |V|+|E| exceeds {t+2 choose 2}. By the proof's reduction, this would require some vertex to have degree greater than 2t−n+1, so checking the maximum degree against Hajnal's bound on any candidate τ-critical graph would settle whether the claim is false.

Watch

Extended reading notes

Core claim

The central result is a sharp inequality: if G=(V,E) is a τ-critical graph of order n with transversal number τ(G)=t, then |V|+|E| ≤ {t+2 choose 2}. The proof derives this by applying Hajnal's degree bound, which states that every vertex in such a graph has degree at most 2t−n+1. Summing this bound over all vertices gives |E| ≤ n(2t−n+1)/2, and the combined quantity |V|+|E| is then at most n + n(2t−n+1)/2. A short algebraic manipulation shows this upper bound is at most {t+2 choose 2} for all n in the feasible range t+1 ≤ n ≤ 2t, with equality only when n=t+1 or n=t+2. The equality cases are then classified: n=t+1 forces a complete graph K_n, while n=t+2 forces the complement to be a single cycle of length at most five, yielding exactly 2K2 and C5.

Load-bearing premise

The proof relies entirely on Hajnal's degree bound (Theorem A), which is quoted without proof; if that bound failed for any τ-critical graph, the degree-sum step and the combined bound would not follow.

Editorial extensions

If this is right

  • The combined bound immediately implies the earlier edge bound |E| ≤ {t+1 choose 2}, because |V| ≥ t+1 in any τ-critical graph.
  • The equality classification gives a complete description of all τ-critical graphs that achieve the combined maximum, which may serve as a starting point for describing near-extremal graphs.
  • Since the proof reduces the whole problem to a single degree-sum argument, any strengthening of Hajnal's bound would automatically produce a corresponding strengthening of the combined inequality.
  • The result closes the loop on the two classical bounds (vertex and edge) by showing they are compatible in a single quadratic inequality with a clean extremal set.
  • The extremal list—complete graphs, two disjoint edges, and a five-cycle—shows that the maximum is attained only by highly regular, highly symmetric examples, suggesting that τ-critical graphs with large order plus size must be tightly structured.

Reading between the lines

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

  • A natural next step beyond the paper is to classify near-equality cases: graphs where |V|+|E| is within an additive constant of {t+2 choose 2}. Based on the proof structure, these should be graphs whose degrees are close to 2t−n+1, likely forcing a long cycle in the complement when n=t+2.
  • The combined bound might extend to other criticality notions, such as edge-critical or matching-critical graphs, where analogous degree bounds are known; the same degree-sum template could yield combined bounds of the form f(t)+g(t).
  • One could test the stability of the result by perturbing τ: if a graph is nearly τ-critical in the sense that τ(G−e) ≥ τ(G)−1 for most edges, does a relaxed version of the bound hold with an error term depending on the number of exceptional edges?
  • The proof's dependence on Hajnal's theorem suggests that a self-contained proof of that degree bound would make the entire argument elementary and may reveal why equality occurs exactly at the three families.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

1 major / 2 minor

Summary. Let G=(V,E) be a τ-critical graph with τ(G)=t. The paper proves the sharp combined bound |V|+|E| ≤ binom(t+2,2), improving the classical separate bounds |V| ≤ 2t (Erdős–Gallai) and |E| ≤ binom(t+1,2) (Erdős–Hajnal–Moon). The proof sums Hajnal's degree bound deg_G(x) ≤ 2t−n+1 over all vertices, then compares the resulting quadratic expression with binom(t+2,2), showing that equality is possible only when n=t+1 or n=t+2. The equality cases are analyzed and the extremal graphs are characterized as the complete graphs K_n (n≥2), 2K2, and C5.

Significance. If the equality-case gap identified below is repaired, this is a clean and useful strengthening of classical results on τ-critical graphs. The derivation is short, essentially parameter-free, and the extremal list is explicit and checkable. The main theorem is falsifiable, and the proof is self-contained modulo the cited Hajnal theorem. However, the printed proof contains an incorrect assertion in the equality characterization that must be corrected before the paper can be accepted; the theorem itself appears true and the repair is straightforward.

major comments (1)
  1. [Proof of Theorem 1, equality case] The proof states: 'In the second case the candidates are the graphs whose complements are 2-regular (and have at least four vertices). Since these graphs are τ-critical with t = n−2, the deletion of any edge creates a set of three vertices inducing no edges.' The universal claim that every graph whose complement is 2-regular is τ-critical is false. For n=6, the complement of C6 (the triangular prism) has τ(G)=4=n−2, but deleting the edge of G corresponding to a distance-3 chord of C6 leaves τ(G−e)=4, not 3, so G is not τ-critical. Consequently, the stated justification of the triangle property is invalid. The desired property does follow directly from the theorem's hypothesis that G is τ-critical together with n=t+2: since τ(G−e)=t−1=n−3, we have α(G−e)=3, which means adding the corresponding chord to the 2-regular complement creates a triangle. Please rewrite this passage so that the triangle property is derived from the τ-criticality of the given graph G rather than from a false assertion about all 2-regular complements; the subsequent cycle-length reasoning is then sound.
minor comments (2)
  1. [Introduction, first paragraph] The phrase 'defined to be the the minimum cardinality' contains a duplicated article; it should read 'defined to be the minimum cardinality'.
  2. [Proof of Theorem 1, inequality (3)] The text says 'with equality if G is a (2t − n + 1)-regular graph'; this should be 'if and only if', because equality in (3) is required for the subsequent conclusion that equality in (2) forces n=t+1 or n=t+2.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the combined bound is derived from an external degree theorem (Hajnal) and the equality cases are analyzed independently; the known equality-case proof gap is a correctness issue, not a circularity.

full rationale

The derivation is self-contained conditional on the cited external Theorem A (Hajnal). The paper obtains |V|+|E| <= n + n(2t-n+1)/2 by summing degree bounds, then reduces the desired bound (2) to (n-t-1)(n-t-2) >= 0 using n >= t+1. No parameter is fitted and the target bound is not assumed. The equality-case proof contains an overgeneralization about complements of 2-regular graphs being tau-critical, but that is a mathematical error in the extremal analysis, not a circularity, because it does not presuppose the theorem or define the conclusion into the hypotheses. Citations to Lovasz and Plummer are background references, not load-bearing. Therefore the circularity score is 0.

Assumptions & free parameters 0 free parameters · 2 assumptions · 0 invented entities

No free parameters or invented entities. The central claim rests on Hajnal's degree bound (Theorem A) taken from the literature, plus the standard observation n ≥ t+1 for τ-critical graphs without isolated vertices.

assumptions (2)
  • domain assumption Hajnal's degree bound: in a τ-critical graph of order n and transversal number t, every vertex has degree at most 2t - n + 1
    Quoted as Theorem A from [3]; the proof of Theorem 1 sums this bound over all vertices to get the edge bound.
  • domain assumption n ≥ t+1 for any τ-critical graph with no isolated vertices
    Used to conclude (n-t-1)(n-t-2) ≥ 0; a graph with n≤t would have too few vertices to have a vertex cover number t while having no isolated vertices.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Order plus size of $\tau$-critical graphs." pith.science (2026). https://pith.science/paper/7ZP4EAA7

@misc{pith2026190805225,
  author       = {Pith},
  title        = {Pith review of: Order plus size of $\tau$-critical graphs},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/7ZP4EAA7}},
  note         = {Machine review of arXiv:1908.05225}
}
abstract

Let $G=(V,E)$ be a $\tau$-critical graph with $\tau(G)=t$. Erd\H{o}s and Gallai proved that $|V|\leq 2t$ and the bound $|E|\leq {t+1\choose 2}$ was obtained by Erd\H{o}s, Hajnal and Moon. We give here the sharp combined bound $|E|+|V|\leq {t+2\choose 2}$ and find all graphs with equality.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

6 extracted references · 6 canonical work pages

  1. [1]

    Erd o s and T

    P. Erd o s and T. Gallai, On the maximal number of vertices representing the edges of a graph, K\"ozl. MTA Mat. Kutat\'o Int. Budapest 6 (1961) 181--203

  2. [2]

    Erd o s, A

    P. Erd o s, A. Hajnal, and J. W. Moon: A problem in graph theory, Amer. Math. Monthly 71 (1964) 1107--1110

  3. [3]

    Hajnal, A theorem on k -saturated graphs, Canadian Journal of Math

    A. Hajnal, A theorem on k -saturated graphs, Canadian Journal of Math. 17 (1965) 720--724

  4. [4]

    Lov\'asz, Combinatorial problems and exercises , Second Edition, AMS Chelsea Publishing, Providence, RI , 2007

    L. Lov\'asz, Combinatorial problems and exercises , Second Edition, AMS Chelsea Publishing, Providence, RI , 2007

  5. [5]

    Lov\'asz, and M.D

    L. Lov\'asz, and M.D. Plummer, Matching Theory . Akad\'emiai Kiad\'o, North Holland 1986

  6. [6]

    Sur\'anyi, On line-critical graphs, in: Infinite and Finite Sets , North Holland, Amsterdam (1975) 1411--1444

    L. Sur\'anyi, On line-critical graphs, in: Infinite and Finite Sets , North Holland, Amsterdam (1975) 1411--1444

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.