{"id":"63b9a23d-ec4c-4661-ae95-512b5f6c4d46","arxiv_id":"math/0602531","paper_version":3,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":8.0,"correctness_risk":"low","formal_verification":"none","parameter_count":0,"one_line_summary":"If a simplicial complex H is a minor of K, then a non-vanishing Van Kampen obstruction in H forces one in K, giving a new non-embeddability criterion.","lead":"The paper extends the idea of graph minors to higher-dimensional simplicial complexes and proves that these higher minors preserve a topological obstruction to embedding. It also settles a question about which edge contractions in triangulated manifolds preserve the shape of the space.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"I have no significant objection: the proof of Theorem 1.1 is internally consistent and the contraction restriction is part of the theorem's stated scope.","rationale":"The central theorem's proof was checked step by step. Lemma 3.1's chain map is well-defined: for F in K'\\K, the sum is nonempty and the boundary computation reduces to Observation 3.2, whose proof is valid exactly because admissibility forbids missing faces containing both identified vertices in the relevant skeleton. Lemma 3.3's transfer of Smith classes follows from the equivariant chain map and the identity on 0-cochains. The reader's weakest assumption about the contraction class is really a scope condition, not a source of error: Theorem 1.1 claims monotonicity for this specific minor relation, and broadening it is not required. The paper is open about the failure for arbitrary identifications. The only real gap is Theorem 4.2's omitted proof details; because the paper labels this an analogue and the central applications use the Z2 version, the omission does not overturn the main claim but does justify a CONDITIONAL verdict. Hence the verdict is unchanged.","tokens_in":13937,"tokens_out":18330,"duration_ms":154536,"concrete_test":"Independently expand the proof of Theorem 4.2: re-derive the claim that the map phi defined in Eq. (8) is an injective Z-chain map and that phi*(om_Z(K x)) = om_Z(K' x) for an admissible contraction, using the explicit evaluation formulas (6)-(7). Apply this to a concrete admissible contraction in a small complex (e.g., identifying two vertices of a 2-sphere triangulation that satisfy the missing-face condition) and check the equality on a 2-dimensional cochain. If the equality fails, Theorem 4.2 is false, though Theorem 1.1 remains unaffected.","verdict_should_be":"UNCHANGED","load_bearing_attack":"I find no load-bearing flaw in the central claim. Theorem 1.1 is proven by an explicit injective chain map for admissible contractions; the construction in Lemma 3.1 is well-defined and injective, and Lemma 3.3 correctly shows that this map pulls back Smith classes. The admissible-contraction condition in the statement of the minor relation is not a hidden assumption: it is the definition under which the transfer property is asserted, and the paper's remark after Theorem 1.1 correctly delineates that broader identifications fail. The genuinely under-supported part is Theorem 4.2, the analogue over Z for deleted products, whose proof omits details; however, that is a secondary result and its omission does not affect the validity of Theorem 1.1 or Corollary 1.2, which depend only on the Z2 Smith class.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper introduces a minor relation for finite simplicial complexes, defined by deletions and \"admissible contractions\" that are allowed only when the two identified vertices are not contained in any missing face of dimension at most dim(K). The main theorem (Theorem 1.1) states that non-vanishing of Van Kampen's obstruction Sm_m (defined via deleted joins and Smith classes with Z2 coefficients) is monotone under this minor relation. The proof constructs an injective chain map from the smaller complex to the larger one and checks that it pulls back the Smith characteristic classes. As a corollary, a complex that contains the d-skeleton of the (2d+2)-simplex as a minor is not embeddable in the 2d-sphere. The paper also proves a PL version of topology-preserving edge contractions (Theorem 1.4), answering a question of Dey et al., and derives lower-bound inequalities for strongly edge decomposable spheres. A second analogue over the integers for deleted products is stated as Theorem 4.2, but its proof is omitted. The paper concludes with examples and open problems comparing higher minors with graph minors.","tokens_in":14108,"tokens_out":15847,"duration_ms":130709,"significance":"The notion of higher minors is a natural and potentially influential generalization of graph minors, and the main monotonicity result provides a useful necessary condition for non-embeddability. The proof of Theorem 1.1 is explicit and self-contained: the chain map in Lemma 3.1 is carefully defined and proved injective, and Lemma 3.3 gives a convincing verification of the Smith-class pullback. The paper is honest about the scope of the result, explicitly noting that arbitrary vertex identifications break the transfer property. Theorem 1.4 is a significant contribution to the edge-contraction literature, with applications to f-vectors. The main weakness is the omitted proof of Theorem 4.2; this does not affect the validity of the central Z2 results, but it is a notable gap in a stated theorem.","major_comments":[],"minor_comments":[{"comment":"The proof of Theorem 4.2 is omitted: the author states that the verification that φ is an injective Z-chain map is similar to Lemma 3.1 and that the equality φ^*(om_Z(K×)) = om_Z(K'×) follows by straightforward checks, but no details are given; since this theorem is used to derive the partial confirmation of Conjecture 1.3, a full proof or a detailed sketch should be provided.","section":"Section 4 (Theorem 4.2)"},{"comment":"In the first paragraph, \"has proved be to very fruitful\" contains a typo; it should read \"has proved to be very fruitful.\"","section":"Section 1 (Introduction)"},{"comment":"In the displayed exact sequence, \"H 0 S(K∗) Sm− →H 1 S(K∗)\" is typeset confusingly: the arrow labeled Sm appears as a minus sign, making it look like an index shift rather than the connecting homomorphism; please use a clear notation such as \\xrightarrow{Sm}.","section":"Section 2 (Algebraic-topological background)"},{"comment":"The proof of Example 6.1 is very terse; in particular, the statement \"by Alexander duality a sphere can not contain a sphere of the same dimension as a proper subspace\" is correct but should be \"cannot,\" and the argument that admissible contractions preserve knottedness could be expanded for readability.","section":"Example 6.1"},{"comment":"In Figure 1, the vertex labels are small and some thin edges are hard to distinguish; please redraw with larger fonts and clearer edge styles.","section":"Figure 1"},{"comment":"The counterexample involving splitting a vertex of K5 is described informally; a precise simplicial description of the complex K would help the reader verify the claim that Sm3(K) = 0 and that identifying u with v yields K5.","section":"Remark after Theorem 1.1"}],"recommendation":"minor_revision","confidential_remarks":"I recommend minor revision. The central Z2 results (Theorem 1.1 and its consequences) are sound and well proved, and the PL edge-contraction theorem is a solid contribution. The main gap is the omitted proof of Theorem 4.2; this should be supplied in a revision, but it does not undermine the paper's central claims. The paper fits the journal's scope and I see no novelty concerns."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nThe short version: this paper generalizes graph minors to simplicial complexes by allowing deletions and 'admissible contractions' (identify u and v only when they are not both contained in any missing face of dimension ≤ dim(K)), and proves that Van Kampen's obstruction in dimension m is monotone under this relation: if H is such a minor of K and Sm_m(H) ≠ 0, then Sm_m(K) ≠ 0. That is a genuine new result, not a routine extension. The restriction on contractions is essential—the remark after Theorem 1.1 shows arbitrary identifications break the property—so it is a definitional choice that does real work, not a hidden assumption.\n\nThe main proof is solid. Lemma 3.1 constructs an explicit injective chain map for admissible contractions, and Lemma 3.3 shows it pulls back Smith classes. The argument is self-contained and uses standard algebraic topology. The Dey et al. question gets a clean affirmative answer: for compact PL manifolds without boundary, an edge contraction is PL-homeomorphic iff the Link Condition holds. The proof, with Newman's theorem and a Mayer-Vietoris sequence, is careful. Corollary 5.3 on nonnegative g-vectors for strongly edge decomposable spheres follows naturally.\n\nWhere are the soft spots? The secondary Theorem 4.2, the analogue over Z for deleted products, states that 'the checks are straightforward' and omits the details. This is a minor gap in a secondary result—the main theorem and the corollaries only need the Z2 case. Still, for a journal submission, those details should be supplied or explicitly referenced. Nothing else rises to the level of a flaw. Example 6.1 is a nice touch, and the paper is honest about where the higher-minor analogy breaks.\n\nWho is this for? Anyone working on embeddability of complexes, Van Kampen obstruction, or graph-minor-style questions in higher dimension. It is a well-written, citable paper. I would send it to a serious referee. The referee should ask for the Theorem 4.2 details, but the central contribution is sound.","headline":"New higher-minor relation with a solid monotonicity theorem for Van Kampen's obstruction; minor gaps in a secondary proof, but the core is sound and deserves peer review.","tokens_in":14550,"tokens_out":1972,"would_cite":true,"duration_ms":18952,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"pith_extraction":{"msc":["05E45","57Q35","05C83"],"pacs":[],"model":"deepseek-v4-flash","headline":"The paper proves that Van Kampen's obstruction is monotone under a new notion of 'higher minor' for simplicial complexes.","keywords":["simplicial complexes","higher minors","Van Kampen obstruction","Smith characteristic class","embeddability","admissible contraction","PL manifolds","g-vector"],"falsifier":"Compute $\\mathrm{Sm}_5(1_{L_*})$ for the complex $L$ of Example 3.6 (a triangulated 4-sphere with a missing triangle added to its 2-skeleton); if it vanishes, then Theorem 1.1 and Corollary 1.2 would be false, whereas the paper asserts it is nonzero.","tokens_in":13777,"feed_emoji":"🧩","tokens_out":13388,"duration_ms":99608,"temperature":0.7,"pith_summary":"The paper defines a notion of minor for finite simplicial complexes that extends graph minors: deletions are ordinary subcomplex removals, and an admissible contraction identifies two vertices only if they are not both contained in any missing face—a set that is not a face while all its proper subsets are—of the complex. The main theorem states that Van Kampen's obstruction—the Smith characteristic class in the equivariant cohomology of the deleted join that detects non-embeddability in a sphere—is monotone under this minor relation: if $H$ is a minor of $K$ and the obstruction of $H$ is nonzero, then the obstruction of $K$ is nonzero. This gives a combinatorial route to non-embeddability: any complex that has the $d$-skeleton of the $(2d+2)$-simplex as a minor cannot be embedded in the $2d$-sphere. The same machinery answers a question about topology-preserving edge contractions in triangulated closed PL manifolds and yields nonnegativity of the $g$-vector for a family of triangulated spheres.","feed_headline":"Higher minors preserve the Van Kampen obstruction","feed_subtitle":"If a complex reduces to a known non-embeddable skeleton, no safe contraction or deletion can make it embeddable.","key_machinery":"The central object is the higher-minor relation: an admissible contraction identifies two vertices $u,v$ only if they are not both contained in any missing face of dimension at most $\\dim(K)$, equivalently $\\operatorname{skel}_{\\dim(K)-2}(\\operatorname{lk}(v,K) \\cap \\operatorname{lk}(u,K)) = \\operatorname{lk}(\\{v,u\\},K)$; a minor is any complex obtained by a sequence of admissible contractions and deletions. The argument is carried by the injective chain map $\\varphi$ of Lemma 3.1, which sends a face $F$ of $K'$ to itself if $F \\in K$, and otherwise to the $\\mathbb{Z}_2$-sum of the faces $(F\\setminus\\{v\\}) \\cup \\{v_0\\}$ obtained by replacing vertices of $F$ by the contracted vertex $v_0$. This map is shown to be a chain map, injective, and to commute with the Smith connecting homomorphisms, so it pulls the fundamental cocycle back correctly and transfers the non-vanishing of the obstruction.","core_discovery":"Theorem 1.1 states that if $H < K$ (H is a higher minor of K) and $\\mathrm{Sm}_m(H) \\neq 0$, then $\\mathrm{Sm}_m(K) \\neq 0$, where $\\mathrm{Sm}_m$ is the $m$-th Smith characteristic class of the deleted join $K_*$—Van Kampen's obstruction. The proof constructs, for every admissible contraction $K \\to K'$, an injective chain map $\\varphi: C_*(K'; \\mathbb{Z}_2) \\to C_*(K; \\mathbb{Z}_2)$ that induces a pullback on equivariant cohomology sending $\\mathrm{Sm}_m(1_{K_*})$ to $\\mathrm{Sm}_m(1_{K'_*})$. Under a deletion the inclusion does the same, so the property survives arbitrary sequences of these moves. Consequently non-embeddability in $\\mathbb{R}^m$ is inherited upward along the higher-minor relation, with the Van Kampen–Flores skeleton serving as the canonical seed obstruction.","pith_inferences":["The monotonicity suggests that the higher-minor relation can be used as a search tool: enumerating admissible contractions and deletions from a candidate complex until a known non-embeddable skeleton appears would certify non-embeddability without computing the obstruction directly.","Because the paper shows that the graph-minor well-quasi-ordering property fails in higher dimensions (Proposition 6.5), this monotonicity is about a single obstruction rather than a complete obstruction theory; a natural extension would be to ask which other characteristic classes behave monotonically under the same moves.","The equivalence between admissibility and PL-homeomorphism preservation indicates that the same contraction rule may serve as a simplification operation in computational topology that rigorously preserves homeomorphism type, potentially simplifying recognition of PL spheres."],"forward_implications":["If the $d$-skeleton of the $(2d+2)$-simplex is a higher minor of $K$, then $K$ is not embeddable in the $2d$-sphere (Corollary 1.2).","An edge in a triangulation of a compact PL manifold without boundary can be contracted to a PL-homeomorphic manifold if and only if it satisfies the Link Condition (Theorem 1.4).","The $g$-vector of every strongly edge decomposable triangulated sphere is nonnegative (Corollary 5.3).","The analogous monotonicity holds for the integer-coefficient Van Kampen obstruction on deleted products (Theorem 4.2), which verifies the embeddability-inheritance conjecture when $2\\dim(H)=m \\neq 4$."],"supporting_citations":[{"why":"Supplies the original Van Kampen result that the relevant skeleton is non-embeddable, the seed for Corollary 1.2.","marker":"[20]"},{"why":"Supplies Flores's independent proof of the non-embeddability of the $d$-skeleton in the $2d$-sphere.","marker":"[5]"},{"why":"Provides the Smith-theoretic formulation of the obstruction used throughout the paper.","marker":"[15]"},{"why":"Gives the non-vanishing and vanishing theorems for the Smith characteristic class that the main proof invokes.","marker":"[16]"},{"why":"Poses the problem about topology-preserving edge contractions that Theorem 1.4 answers.","marker":"[3]"},{"why":"Supplies the PL topology facts (balls, stars, Newman's theorem) used in the proof of Theorem 1.4.","marker":"[7]"},{"why":"Provides background for the deleted-product obstruction over $\\mathbb{Z}$ used in Section 4.","marker":"[23]"}],"fun_headline_variants":["Van Kampen obstruction persists under higher minors","Higher minors cannot erase non-embeddability","Minor moves preserve Van Kampen obstruction","Non-embeddability survives higher minors"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The proof relies on the delicate definition of an admissible contraction, which forbids identifying two vertices that lie together in a missing face of the complex; if the contraction rule were broadened to arbitrary vertex identifications, the monotonicity statement would be false.","fun_headline_variants_meta":{"raw":{"variants":["Van Kampen obstruction persists under higher minors","Higher minors cannot erase non-embeddability","Minor moves preserve Van Kampen obstruction","Non-embeddability survives higher minors"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00024,"raw_usage":{"total_tokens":1500,"prompt_tokens":911,"completion_tokens":589,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":527,"completion_tokens_details":{"reasoning_tokens":534}},"tokens_in":527,"tokens_out":589,"duration_ms":5134,"temperature":1.0,"reasoning_tokens":534,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-28T20:00:45.173585+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute $\\mathrm{Sm}_5(1_{L_*})$ for the complex $L$ of Example 3.6 (a triangulated 4-sphere with a missing triangle added to its 2-skeleton); if it vanishes, then Theorem 1.1 and Corollary 1.2 would be false, whereas the paper asserts it is nonzero.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the original Van Kampen result that the relevant skeleton is non-embeddable, the seed for Corollary 1.2."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies Flores's independent proof of the non-embeddability of the $d$-skeleton in the $2d$-sphere."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the Smith-theoretic formulation of the obstruction used throughout the paper."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Gives the non-vanishing and vanishing theorems for the Smith characteristic class that the main proof invokes."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Poses the problem about topology-preserving edge contractions that Theorem 1.4 answers."},{"cited_title":"Hudson, Piecewise-linear topology, Benjamin Inc., New York 1969","cited_arxiv_id":null,"evidence_quote":"Supplies the PL topology facts (balls, stars, Newman's theorem) used in the proof of Theorem 1.4."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides background for the deleted-product obstruction over $\\mathbb{Z}$ used in Section 4."}],"review_version":1}