Bricks that every removable edge is solitary

Fuliang Lu; Jinxin Xue; Jun Ge; Yaxian Zhang

REVIEW 4 major objections 6 minor 20 references

Bricks that every removable edge is solitary

T0 review · 4 major / 6 minor · reviewed 2026-08-15 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Every nonsolid brick with all removable edges solitary is a splicing of an odd wheel and a smaller brick.

desk verdict Plausible theorem, but the main proof has a load-bearing gap: G/X' is not H⊙G2, and Lemma 2.13 is applied to the wrong contraction. read the letter →

arxiv 2608.12832 v1 pith:POO7EH5C submitted 2026-08-13 math.CO

Jinxin Xue , Jun Ge , Fuliang Lu , Yaxian Zhang This is my paper · ORCID

classification math.CO MSC 05C7005C4005C75

keywords brickgraphperfectmatchingremovableedgesolitarycoveredtightcutdecompositionsplicingoddwheel

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

Bricks are 3-connected graphs that lose a perfect matching only when any two vertices are deleted. An edge is removable if deleting it keeps the graph matching covered, and solitary if it belongs to exactly one perfect matching. This paper proves that every simple nonsolid brick whose removable edges are all solitary is a splicing of an odd wheel, up to multiple edges, with a smaller brick that also has the property. This strengthens an earlier open problem about b-invariant solitary edges, and it gives a recursive recipe: starting from such smaller bricks and repeatedly splicing odd wheels builds the whole nonsolid portion of the class. The paper also constructs an infinite family showing the odd wheel in this recipe cannot be replaced by K4.

What carries the argument

The mechanism is splicing together with the robust cut decomposition of nonsolid bricks. A splicing of two graphs removes a vertex of equal degree from each and identifies the dangling edges along a bijection; the paper uses the fact that splicing matching covered graphs is matching covered, with a criterion for when the result is a brick. The proof starts from a robust cut, a separating cut whose two contractions are near-bricks, refined by contracting subsets so that one side is a solid brick, the other is a brick, and the contracted remainder H is bipartite and matching covered. It then forces the solid side to be an odd wheel and forces H to have exactly two vertices, so the only remaining structure is the splice.

What would settle it

Check the finite catalogue of simple nonsolid bricks up to, say, twelve vertices: if any brick with every removable edge solitary cannot be written as a splicing of an odd wheel and a smaller brick with the same property, Theorem 1.2 is false. A concrete way to look is to run the robust-cut construction on each candidate and test whether the contracted solid side is always an odd wheel, which the proof says it must be.

Watch

Extended reading notes

Core claim

The central claim, Theorem 1.2, is that a simple nonsolid brick G in which every removable edge is solitary decomposes as a splicing of an odd wheel and a brick G2 whose removable edges are also all solitary. The proof selects a robust cut of G from a known structural lemma; one side contracts to a solid brick G1 whose underlying simple graph is forced to be an odd wheel, while the other contracts to G2. The key step shows the fully contracted middle graph H has only the two contracted vertices, so the original brick is exactly the splice of G1 and G2. Consequently every nonsolid brick in the class can be built by repeatedly splicing odd wheels into bricks of the same kind.

Load-bearing premise

The theorem's proof leans on a cited structural lemma that every nonsolid brick admits a robust cut with refinements making one contraction a solid brick, the other a brick, and the fully contracted remainder bipartite and matching covered; if that lemma has a counterexample, the decomposition claimed in Theorem 1.2 is not established.

Editorial extensions

If this is right

Every simple nonsolid brick with all removable edges solitary is obtained from smaller bricks of the same kind by finitely many splices of odd wheels, up to multiple edges.
The odd wheel in the decomposition cannot be replaced by K4: the infinite family Gi consists of nonsolid bricks with all removable edges solitary, none of which is a splice of a brick and K4 up to multiple edges.
In any splice of two odd wheels that is a brick with all removable edges solitary, at least one wheel must be K4, and if the other wheel has at least five rim vertices, the splicing vertex cannot be that wheel's hub.
If a brick built by splicing has removable edges equal to its solitary edges, that equality is inherited from the two factors; conversely, the spliced brick has removable edges contained in solitary edges only when the same containment already holds in the factors.

Reading between the lines

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

A complete classification would follow if the same odd-wheel conclusion held for solid bricks satisfying the property: Corollary 2.16 already forces any solid side that appears in the robust-cut argument to be an odd wheel, so a matching theorem for solid bricks would turn the one-step decomposition into a full recursive generation from odd wheels alone.
The layer-gadget construction suggests how to build test families with prescribed high-degree vertices; varying the terminal wheels and the number of layers may give extremal examples for deciding whether the decomposition is unique or whether splice order matters.
The same machinery, robust cuts, contracting to a bipartite matching-covered remainder, then showing the remainder has two vertices, is likely to transfer to other hereditary edge properties such as 'every removable edge is b-invariant,' provided an analogue of the non-solitary-edge lemma holds for the inherited property.

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.

Desk Editor's Note

read the letter

The main theorem is plausible, but the proof as written does not go through. The problem is in Claim 1, where the paper asserts G/X' := H⊙G2. Under the paper's own definition, splicing requires vertex-disjoint graphs, and H (which contracts X' and X'' to x', x'') and G2 = G/X'' are not disjoint: they share x'' and all vertices outside X'∪X''. So the splicing H⊙G2 isn't defined, and taking disjoint copies gives the wrong vertex count. The subsequent use of Lemma 2.5 also doesn't work, because the two C-contractions of G/X' for C=∂(X'') are H and the graph obtained by contracting everything outside X'' to one vertex—not H and G2. Separately, Lemma 2.13, as stated, concludes that every removable edge of the other C-contraction is solitary; the paper instead concludes it about G/X', the wrong contraction unless notation is clarified. So the central claim that V(H)={x',x''} is unproved, and without it the splicing decomposition is not established.

There are smaller gaps too: the 'it can be checked' statements in Section 3.2 for the constructed family, and an unproved assertion in the proof of Lemma 2.13 that a removable doubleton is a perfect matching (it is a pair of edges, not a perfect matching in general).

What's good: the paper attacks a natural strengthening of the Lucchesi–Murty problem, extends prior solid-brick results to a class of nonsolid bricks, and the idea of using robust cuts and the solid-brick lemma is a sensible route. The construction showing K4 can't replace odd wheels is potentially interesting. The theorem may well be true, but this version does not establish it.

I'd still send it to a referee: the topic is squarely within matching theory, the authors have relevant prior work, and the gap is identifiable and possibly fixable. But I'd ask the referee to check Claim 1 and the applications of Lemma 2.5 and Lemma 2.13 carefully. I wouldn't cite it until the proof is repaired.

Referee Report

4 major / 6 minor

Summary. The paper studies bricks (3-connected bicritical graphs) in which every removable edge is solitary, i.e., belongs to a unique perfect matching. The main result, Theorem 1.2, states that every simple nonsolid brick with this property is a splicing of an odd wheel (up to multiple edges) and a brick in which every removable edge is solitary. The proof uses a structural lemma (Lemma 2.7) to find a robust cut with two contracted subgraphs G1 and G2 and a bipartite matching covered graph H. The authors then attempt to show |V(H)|=2 via Claim 1, which would decompose G as a splice. Section 3.2 constructs an infinite family of bricks satisfying the property that cannot be obtained by splicing with K4, and Section 4 discusses bricks where the sets of removable and solitary edges coincide.

Significance. If correct, Theorem 1.2 gives a recursive generation of all nonsolid bricks whose removable edges are all solitary, strengthening recent results on b-invariant edges and the Lucchesi-Murty problem. The construction of the infinite family in Theorem 3.1 is interesting, and the results on R=S in Section 4 are useful. However, the proof of Theorem 1.2 contains a load-bearing error in Claim 1, and the verification of Theorem 3.1 relies on unchecked assertions; these issues must be resolved before the results can be accepted. The paper makes good use of external structural theorems (Tutte, tight cut decomposition, robust cut lemma) and cites the relevant literature.

major comments (4)

[Section 3.1, Claim 1] The asserted identity 'G/X′ := H⊙G2' is false. Under the splicing definition in Section 2.1, the two graphs being spliced must be vertex-disjoint, but H and G2 share all vertices outside X′∪X″; moreover, H contains x″ as a single vertex while G2 contains the individual vertices of X′ and the same outside vertices. Consequently the vertex count of a splice of H and G2 is 2|V(G)|−|X′|−2|X″|+1, whereas |V(G/X′)|=|V(G)|−|X′|+1, so equality cannot hold in general. This invalidates the claim that ∂(X′) is separating and the subsequent use of Lemma 2.13 to conclude that every removable edge of G/X′ is solitary. The proof of Claim 1, and hence Theorem 1.2, does not go through as written.
[Section 3.1, application of Lemma 2.13] Even if the splicing identity were corrected, the sentence 'By Lemma 2.13, every removable edge of G/X′ is solitary' is not justified. Lemma 2.13 concludes that every removable edge of the other C-contraction is solitary, so if G1=G/X′ is the odd wheel, the conclusion would apply to the complementary contraction, not to G/X′ itself. If the intention is that H plays the role of the odd wheel, then H is bipartite and cannot be an odd wheel. The logical structure of this step needs to be clarified and repaired.
[Section 3.2, Claim A of Theorem 3.1] The proof of Claim A states 'It can be checked that Mi \ Si = E(Gi) \ Si' and asserts without argument that Si is exactly the set of all solitary edges of Gi. These assertions are the core of the verification that every removable edge of Gi is solitary; leaving them as unchecked checkable claims is not a formal proof. A detailed argument, or a verifiable enumeration for the construction, is required for Theorem 3.1 to support the paper's advertised conclusion about K4.
[Lemma 2.13, case k=3] The sentence 'Obviously, R is also a perfect matching of G1' requires proof, since a removable doubleton is not in general a perfect matching in an arbitrary brick. If this fact holds for K4 (possibly with multiple edges), it should be proved or cited; as written, the argument that the removable doubleton R serves as a perfect matching is unsupported.

minor comments (6)

[Figure 1] The caption of Figure 1 contains garbled labels such as '1/g16kw' and 'w1/g99'; the vertex labels in the figure should be corrected.
[Section 3.1, after Claim 1] The notation G2 is redefined: earlier G2=G/X″, and later 'Then G2 := G/X′'. This is confusing and should be clarified by using distinct names for the two contractions, for example by writing the two C-contractions explicitly.
[Section 3.2, Theorem 3.1] The proof uses the phrases 'It can be checked' and 'Obviously' repeatedly; these should be replaced by explicit arguments or by a reference to a figure with a verifiable enumeration of edges and perfect matchings.
[Proposition 3.2] The notation 'Ws(x) ⊙ Wt(x)' uses the same symbol x for the splicing vertex in both wheels, which is ambiguous; different labels should be used for the two splicing vertices.
[Section 4, Figures 2 and 3] The figures are described as using red edges, but color may not be visible in print; the captions should describe the distinction between edge types textually as well.
[Introduction and abstract] The paper asserts that every b-invariant edge is removable; this is used in the motivation but is not proved or cited. A brief justification or reference would make the strengthening explicit.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the proof is self-contained modulo external structural lemmas.

full rationale

The paper's main derivation is not circular. Theorem 1.2 is proved by importing structural decomposition facts (Lemma 2.7 from [8], Lemmas 2.14/2.15 from [14]/[6], Lemma 2.9 from [6]) and by proving auxiliary lemmas (2.11-2.13, 4.1, 4.2) from the definitions of matching covered graphs, removable edges, and solitary edges. The cited results have hypotheses about robust cuts, solid bricks, or forcing edges that do not contain the theorem's conclusion about splicing odd wheels; they are external structural inputs rather than restatements of the target result. No parameter is fitted and no quantity called a prediction is taken from the data. The suspicious identity G/X' := H⊙G2 in Claim 1 would be a proof defect if false, but a false inference is a correctness issue, not a circular reduction of the theorem to its own assumption. Consequently there is no self-definitional, fitted-input, or self-citation-load-bearing circularity.

Assumptions & free parameters 0 free parameters · 4 assumptions · 0 invented entities

No free parameters appear; the paper is pure structural graph theory. The central claim rests on standard theorems (Tutte, Lovasz) and on two external structural lemmas from papers with overlapping authorship (robust cut lemma [8], solid brick lemma [6]). The constructed family {G_i} is a mathematical example, not a postulated entity.

assumptions (4)

standard math Tutte's 1-factor theorem
Invoked as Theorem 2.1 and used to derive Proposition 2.2 on separating cuts in matching covered graphs.
standard math Lovasz tight cut decomposition theorem
Defines bricks, braces, and b(G); provides the background for all notions of removable and b-invariant edges.
domain assumption Lemma 2.7: every nonsolid brick has a robust cut with stated contraction properties
External theorem from He-Lu-Xue [8], a paper with overlapping authorship. The main proof of Theorem 1.2 depends directly on this decomposition.
domain assumption Lemma 2.15: simple solid brick with at most one vertex not incident to a solitary edge is an odd wheel
External theorem from Dai-Lu-Zhang [6]; used to derive Corollary 2.16 which identifies the solid contraction as an odd wheel.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Bricks that every removable edge is solitary." pith.science (2026). https://pith.science/paper/POO7EH5C

@misc{pith2026260812832,
  author       = {Pith},
  title        = {Pith review of: Bricks that every removable edge is solitary},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/POO7EH5C}},
  note         = {Machine review of arXiv:2608.12832}
}

abstract

A brick is a 3-connected graph $G$ such that $G-u-v$ has a perfect matching for any two distinct vertices $u,v\in V(G)$. An edge $e$ in a matching covered graph $G$ is removable if $G-e$ is matching covered. We say that a removable edge $e$ in a brick $G$ is $b$-invariant if $b(G-e)=b(G)=1$, where $b(H)$ denotes the number of bricks in the tight cut decomposition of a matching covered graph $H$. An edge of a graph is solitary if it lies in precisely one perfect matching. Lucchesi and Murty proposed the problem of characterizing bricks, distinct from $K_4$, $\overline{C_6}$ and the Petersen graph, in which every $b$-invariant edge is solitary. Note that every $b$-invariant edge is removable. In this paper, we strengthen the condition by requiring that every removable edge is solitary. We show that every nonsolid brick satisfying this strengthened condition can be obtained by repeatedly splicing odd wheels (up to multiple edges). Moreover, properties of such bricks imply that "repeatedly splicing odd wheels" cannot be replaced by "repeatedly splicing copies of $K_4$".

Figures

Figures reproduced from arXiv: 2608.12832 by the authors.

**Figure 1.** The graphs G1 and G2. Theorem 3.1. Every graph G in G is a nonsolid brick in which every removable edge of G is solitary. Furthermore, G cannot be obtained by splicing a brick and K4, up to multiple edges. Proof. Let G := Gi be a graph in G, for i ≥ 0. Note that the graph Gi is isomorphic to a graph obtained by sequentially splicing another wheel graph to the initial wheel graph W′ ℓ , where each splicing operation … view at source ↗

**Figure 2.** The graphs G1, G2 and G = (G1(u) ⊙ G2(v))θ. H H1 H2 x y [PITH_FULL_IMAGE:figures/full_fig_p011_2.png] view at source ↗

**Figure 3.** The graphs H1, H2 and H = (H1(x) ⊙ H2(y))θ. References [1] J. A. Bondy and U. S. R. Murty. Graph Theory. Springer, 2008. [2] M. H. Carvalho, C. L. Lucchesi and U. S. R. Murty. Ear decompositions of matching covered graphs. Combinatorica, 19: 151-174, 1999. [3] M. H. Carvalho, C. L. Lucchesi and U. S. R. Murty. On a conjecture of Lov´asz concerning bricks. II. Bricks of finite characteristic. J. Combin. Theory Ser. … view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

20 extracted references · 15 canonical work pages

[1]

J. A. Bondy and U. S. R. Murty. Graph Theory. Springer, 2008

2008
[2]

M. H. Carvalho, C. L. Lucchesi and U. S. R. Murty. Ear decompo sitions of matching covered graphs. Combinatorica, 19: 151-174, 1999

work page 1999
[3]

M. H. Carvalho, C. L. Lucchesi and U. S. R. Murty. On a conject ure of Lov´ asz concern- ing bricks. II. Bricks of ﬁnite characteristic. J. Combin. Theory Se r. B, 85: 137-180, 2002. 11

work page 2002
[4]

M. H. Carvalho, C. L. Lucchesi and U. S. R. Murty. Graphs with in dependent perfect matchings. J. Graph Theory, 48: 19-50, 2005

work page 2005
[5]

M. H. Carvalho, C. L. Lucchesi and U. S. R. Murty. A generalizat ion of Little’s theorem on Pfaﬃan graphs. J. Combin. Theory Ser. B, 102: 1241-1266, 20 12

work page
[6]

X. Dai, F. Lu and Y. Zhang. Bricks in which every vertex is incident w ith a forcing edge. arXiv:2606.26594, 2026

work page Pith review arXiv 2026
[7]

Edmonds, L

J. Edmonds, L. Lov´ asz and W. R. Pulleyblank. Brick decompositions and the matching rank of graphs. Combinatorica, 2(3): 247-274, 1982

work page 1982
[8]

X. He, F. Lu and J. Xue. Wheel-like bricks and minimal matching cove red graphs. J. Graph Theory, 111(1): 5-16, 2026

work page 2026

Show all 20 references

[9]

Kothari, M

N. Kothari, M. H. Carvalho, C. L. Lucchesi and C. H. C. Little. On essentially 4-edge- connected cubic bricks. Electron. J. Combin., 27(1): #P1.22, 2020

2020
[10]

Lov´ asz

L. Lov´ asz. Ear decompositions of matching covered graphs. Combinatorica, 2: 105- 117, 1983

1983
[11]

Lov´ asz

L. Lov´ asz. Matching structure and the matching lattice. J. C ombin. Theory Ser. B, 43: 187-222, 1987

1987
[12]

Lov´ asz and M

L. Lov´ asz and M. D. Plummer. Matching Theory. Number 29 in An nals of Discrete Mathematics. Elsevier Science, 1986

1986
[13]

Lu and J

F. Lu and J. Xue. Planar wheel-like bricks. arXiv:2410.20692, 202 4

arXiv
[14]

C. L. Lucchesi and U. S. R. Murty. Perfect Matchings: A Theo ry of Matching Covered Graphs. Vol. 31. Springer, 2024

2024
[15]

W. T. Tutte. The factorization of linear graphs. J. Lond. Math . Soc., 22: 107-111, 1947

1947
[16]

X. Wu, F. Lu and L. Zhang. Removable edges in claw-free bricks. Graphs Combin., 40: 43, 2024

2024
[17]

Zhang, F

Y. Zhang, F. Lu and H. Zhang. Cubic bricks that every b-invariant edge is forcing. arXiv:2411.17295, 2024

2024 arXiv
[18]

Zhang and X

Y. Zhang and X. Wang. Solid bricks that every b-invariant edge is solitary. arXiv:2507.21565, 2025

2025 arXiv
[19]

Zhang and X

Y. Zhang and X. Wang. Claw-free bricks that every b-invariant edge is solitary. arXiv:2509.23114, 2025

2025
[20]

J. Zhou, X. Feng and W. Yan. Claw-free solid bricks. Graphs and Combinatorics, 41: 86, 2025. 12

2025

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.

[1] [1]

J. A. Bondy and U. S. R. Murty. Graph Theory. Springer, 2008

2008

[2] [2]

M. H. Carvalho, C. L. Lucchesi and U. S. R. Murty. Ear decompo sitions of matching covered graphs. Combinatorica, 19: 151-174, 1999

work page 1999

[3] [3]

M. H. Carvalho, C. L. Lucchesi and U. S. R. Murty. On a conject ure of Lov´ asz concern- ing bricks. II. Bricks of ﬁnite characteristic. J. Combin. Theory Se r. B, 85: 137-180, 2002. 11

work page 2002

[4] [4]

M. H. Carvalho, C. L. Lucchesi and U. S. R. Murty. Graphs with in dependent perfect matchings. J. Graph Theory, 48: 19-50, 2005

work page 2005

[5] [5]

M. H. Carvalho, C. L. Lucchesi and U. S. R. Murty. A generalizat ion of Little’s theorem on Pfaﬃan graphs. J. Combin. Theory Ser. B, 102: 1241-1266, 20 12

work page

[6] [6]

X. Dai, F. Lu and Y. Zhang. Bricks in which every vertex is incident w ith a forcing edge. arXiv:2606.26594, 2026

work page Pith review arXiv 2026

[7] [7]

Edmonds, L

J. Edmonds, L. Lov´ asz and W. R. Pulleyblank. Brick decompositions and the matching rank of graphs. Combinatorica, 2(3): 247-274, 1982

work page 1982

[8] [8]

X. He, F. Lu and J. Xue. Wheel-like bricks and minimal matching cove red graphs. J. Graph Theory, 111(1): 5-16, 2026

work page 2026