{"id":"d5f41e4f-b760-414e-b437-4bd17b64335e","arxiv_id":"2509.07247","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"The authors prove that any point configuration whose convex-hull intersection combinatorics is captured by a Radon pair yields a Fan-type covering or labeling theorem for the sphere, and derive colorful, continuous, and (Z/2)^2-equivariant versions.","lead":"This paper proves a family of generalizations of Fan's combinatorial Borsuk-Ulam theorem for sphere triangulations, replacing fixed cyclic sign patterns with arbitrary Radon-type intersection patterns of point sets. Generalists may care because the framework converts a single topological existence theorem into a richer structural toolkit for coloring, covering, matching, and mass-partition problems.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Theorem 2.5's proof relies on a partition of unity subordinate to the closed cover q(A_i), which can fail for valid covers; the proof as written has a concrete gap.","rationale":"The reader's verdict is CONDITIONAL and already flags the partition-of-unity regularity issue as part of the weakest assumption. My pass isolates it as the single most load-bearing concern because it is not merely an omitted detail: for admissible covers there provably is no such partition of unity, so the proof of Theorem 2.5 is incomplete. This does not refute the paper's central Theorem 1.2, whose fixed-point/limiting proof is sound, and the gap is likely repairable (e.g., by thickening the A_i to open sets and taking a limit). Hence I do not move the verdict to reject; the manuscript remains conditional on supplying a correct proof of the continuous extension. I did not find a comparable flaw in Theorem 1.2 itself, and the other reader-flagged issues (fine triangulation, Ramos's theorem) are less central or more standard.","tokens_in":26634,"tokens_out":35026,"duration_ms":412143,"concrete_test":"Verify the counterexample: on S^1, for 0<a<π/2 set A_1={0}, A_2=[0,a], A_3=[-a,0], A_4=[a,π-a]. Check all hypotheses of Theorem 2.5 (closed sets, A_i∩-A_i=∅, ∪(A_i∪-A_i)=S^1). Then attempt to build α_i on RP^1 as in the proof. At q(0), continuity forces every α_i to vanish because q(0) is an endpoint of q(A_2) and q(A_3), q(0)∉q(A_4), and q(A_1) is a singleton. This yields 0=∑α_i(0)≠1, showing the partition-of-unity step is impossible for this cover and no proof of Theorem 2.5 can proceed along these lines without additional hypotheses or a different construction.","verdict_should_be":"UNCHANGED","load_bearing_attack":"In the proof of Theorem 2.5, the authors choose continuous functions α_i: RP^d→[0,1] with ∑α_i=1 and α_i>0 only on q(A_i). This is asserted via [59], but q(A_i) are closed, and for a closed cover such a partition of unity need not exist. Example on S^1: take 0<a<π/2, A_1={0}, A_2=[0,a], A_3=[-a,0], A_4=[a,π-a]. Then A_i∩(-A_i)=∅ and ∪(A_i∪-A_i)=S^1. In RP^1, q(A_2)=[0,a], q(A_3)=[π-a,π]∪{0}, q(A_4)=[a,π-a], and q(A_1)={0}. The point 0 is an endpoint of q(A_2) and q(A_3) and is not in the interior of any q(A_i). Any continuous α_i that is positive at 0 would be positive in a full neighborhood of 0 in RP^1, but every such neighborhood contains points outside q(A_i); hence α_i(0)=0 for all i, contradicting ∑α_i(0)=1. Thus the required partition of unity does not exist for a perfectly admissible covering. This invalidates the proof of Theorem 2.5 as written; the theorem may be true and repairable by an approximation argument, but the manuscript does not supply it.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proves several generalizations of Ky Fan's combinatorial labeling theorem, which itself generalizes Borsuk–Ulam. Theorem 1.2 asserts that for a point set X={x_1,...,x_m} in R^{d-1} and an antipodal closed cover A_1,...,A_m,-A_1,...,-A_m of S^d with A_i∩(-A_i)=∅, there is a Radon pair (S,T) for X such that ∩_{i∈S} A_i ∩ ∩_{i∈T} (-A_i) is nonempty. This is proved by a short reduction to Borsuk-Ulam via a Radon lemma and a limiting argument. Theorem 2.2 gives a parity statement for generic odd labelings of antipodal triangulations; Theorem 2.5 replaces convex hulls by continuous images h(σ),h(τ) of faces of the simplex; Theorem 2.6 gives a colorful extension. The paper then develops applications to sphere coverings, non-embeddability, Kneser colorings, rainbow faces, the topological Hall theorem, hypergraph Hall theorems, ham-sandwich-type mass partition results, and a (Z/2)^2-equivariant product-sphere version. The central claim is that the sign/combinatorial patterns in Fan-type results are governed by order types of point sets, or more generally by intersection combinatorics of continuous images.","tokens_in":26995,"tokens_out":26180,"duration_ms":292375,"significance":"If the results hold, this is a substantial and useful unification: it shows that the classical alternating-sign pattern in Fan's theorem is just one order type of m points in R^{d-1}, and that every Radon-type intersection pattern yields a Fan-type covering theorem. The proof of Theorem 1.2 is elegant and genuinely short, reducing directly to Borsuk-Ulam through the standard Radon lemma. The colorful versions and the applications to topological Hall theorems and mass partitions give the paper broad reach. The paper is also honest about its dependence on prior work in special cases and on Ramos's theorem for powers of two in the product-sphere section. The main weakness is a concrete gap in the proof of Theorem 2.5, which is one of the advertised main extensions and is used in later sections; this requires repair before the paper can be accepted. The remainder of the core theorems are mostly standard and sound modulo local details.","major_comments":[{"comment":"The proof requires a continuous partition of unity α_i: RP^d→[0,1] with α_i(x)>0 only when x∈q(A_i), citing [59]. Partitions of unity are normally subordinate to open covers, and for a closed cover such a partition need not exist. Concrete example: on S^1 (angles modulo 2π) take 0<a<π/2 and closed arcs A_1={0}, A_2=[0,a], A_3=[-a,0], A_4=[a,π-a]. These satisfy A_i∩(-A_i)=∅ and ∪(A_i∪-A_i)=S^1. In RP^1 the point 0 lies in q(A_1) and is an endpoint of both q(A_2) and q(A_3). Any continuous α_i positive at 0 would be positive in a full neighborhood of 0, but every such neighborhood contains points outside q(A_i); hence α_i(0)=0 for all i, contradicting ∑α_i=1. Thus the proof of Theorem 2.5 does not go through. Since Theorem 2.5 is used in §3.2 and in Theorem 7.1, this is a load-bearing gap. The theorem may well be true and repairable by a triangulation/limiting argument along the lines of T","section":"§2, proof of Theorem 2.5"},{"comment":"The proof asserts that for a signed label set eA with |a| set A, Σ[f^{-1}(A)] is isomorphic to Λ(Σ,f)[ e f^{-1}(eA)]. This is false when eA contains both positive and negative signs. In that case the induced subcomplex is the join of the subcomplexes on the positive-sign and negative-sign labels, not the single complex Σ[f^{-1}(A)]. The connectivity needed for the skeleton extension still follows from the join connectivity theorem, so the statement is repairable, but as written the proof of Lemma 4.9 has an incorrect step. Since this lemma is the route to the claimed new proof of the topological Hall theorem (Theorem 4.4), the gap should be fixed.","section":"§4.1, proof of Lemma 4.9"}],"minor_comments":[{"comment":"The indexing is off by one: vertices of the barycentric subdivision correspond to faces of dimensions 0,...,d, but the sets X^{(j)} and A^{(j)}_i are indexed by j∈[d+1]. The proof writes f(v_σ)=(x^{(dim σ)}_i,1); this should be dim σ+1 (or the index set should be {0,...,d}).","section":"§2, Theorem 2.6 proof"},{"comment":"In the definition of A_i, the expression max_{j∈[m]} |f_j(x0)| uses x0 before it is introduced; it should be max_{j∈[m]} |f_j(x)|.","section":"§3.3, Corollary 3.7"},{"comment":"The two items in Remark 2.7 are both numbered (1); the second should be (2).","section":"§2, Remark 2.7"},{"comment":"The degree argument is terse: e f is defined as a map to R^{d+1}, not to S^d. The proof should explicitly normalize to S^d (the map is nonzero) and justify that the preimage of e_{d+1} is finite before summing local degrees. This is standard and fixable, but as written it jumps a step.","section":"§2, Theorem 2.2 proof"},{"comment":"The proof of Theorem 6.2 assumes the existence of arbitrarily fine (Z/2)^2-symmetric triangulations of S^d×S^{d-1} with small facets. This is plausible and standard, but a one-sentence justification or reference would improve clarity.","section":"§6"}],"recommendation":"major_revision","confidential_remarks":"The paper has a strong core: Theorem 1.2 and its colorful and parity variants are elegant and mostly solid. The serious issue is the partition-of-unity step in Theorem 2.5, which is a genuine gap in a stated main theorem and propagates to Section 7's characterization and to the non-embeddability examples. The false isomorphism in Lemma 4.9 is also embarrassing but easily repairable. I would send back for major revision rather than reject, since the main combinatorial results and the broad applications are likely salvageable with a modest amount of new proof. I would ask the authors specifically to either replace the partition-of-unity argument in Theorem 2.5 with the fine-triangulation/limit method used in Theorem 1.2, or give a correct proof of the existence of the required α_i under the stated hypotheses, and to rewrite the proof of Lemma 4.9 using the join connectivity theorem."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Here's my take on Frick–Wellner. The headline: this paper is worth serious attention, but it has a real hole in one of its advertised main results. Theorem 1.2 is the real deal. For any m points in R^{d-1}, any antipodal sphere cover by closed sets A_i with A_i ∩ (-A_i)=∅ gives disjoint S,T with intersecting convex hulls and the corresponding intersection of A_i's and -A_i's. The proof is a short, elegant reduction to Borsuk–Ulam via the Radon lemma, and Fan's theorem drops out as the special case where the points lie on the moment curve. That alone is a nice unification. The parity result (Theorem 2.2) is fine, a standard degree argument. The applications are extensive: colorful Fan, local LS-type results, a new proof of the topological Hall theorem, ham sandwich variants. I believe the main theorem and most applications are sound.\n\nBut the stress-test caught something real. In the proof of Theorem 2.5, the authors invoke [59] for a partition of unity α_i on RP^d subordinate to the closed cover q(A_1),…,q(A_m), meaning α_i(x)>0 only if x∈q(A_i). For closed covers, such a partition of unity need not exist. The stress-test's S^1 example is correct: at a point 0 that is an endpoint of several q(A_i)'s, any continuous α_i positive at 0 would be positive in a neighborhood of 0, which would force it outside the closed set. So all α_i(0)=0, contradicting Σα_i=1. The proof as written fails at that step. The theorem may well be true—an approximation argument with slightly enlarged open sets or a fine triangulation likely repairs it—but the manuscript doesn't supply the repair. That is a load-bearing gap in a central theorem, not a minor typo.\n\nThe (Z/2)^2 section is also more sketched than the rest, relying on Ramos's theorem for d a power of two and abbreviating the fine-triangulation step. Those are less serious.\n\nOverall: the core insight is good, the writing is clear, and the main theorem deserves citation. But Theorem 2.5 needs a corrected proof before the paper is in final form. I'd send it to a serious referee, not desk reject, and the referee should focus on Section 2.5 and Section 6.","headline":"A genuinely new and clean generalization of Fan's theorem (Theorem 1.2), but the continuous version (Theorem 2.5) has a real proof gap in the partition-of-unity step that the authors need to repair.","tokens_in":27478,"tokens_out":2058,"would_cite":true,"duration_ms":24220,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["55M20","52A35","05C15"],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper proves that the intersection patterns forced by the Borsuk–Ulam theorem are governed by Radon pairs of finite point sets, and extends that principle to continuous, colorful, parity, and product-sphere settings.","keywords":["Borsuk–Ulam theorem","Fan's lemma","Radon partitions","sphere coverings","topological Hall theorem","ham sandwich theorem","local chromatic number","mass partitions"],"falsifier":"Set d=2 and take X={x1,x2,x3,x4} with x4 strictly inside the triangle x1x2x3, so the unique Radon pair is {4} against {1,2,3}. Try to construct closed sets A1,...,A4⊂S^2 with A_i∩(-A_i)=∅ and ∪(A_i∪(-A_i))=S^2 but with A_4∩(-A_1∩-A_2∩-A_3)=∅. Theorem 1.2 says no such covering exists; an explicit covering or a computational search showing one exists would refute it.","tokens_in":26544,"feed_emoji":"🌐","tokens_out":10199,"duration_ms":117558,"temperature":0.7,"pith_summary":"This paper generalizes Ky Fan's combinatorial version of the Borsuk–Ulam theorem. The central result, Theorem 1.2, says: whenever closed sets A_1,...,A_m on the d-sphere each miss their antipodal image and together with their antipodes cover the sphere, and whenever X={x_1,...,x_m} is any point set in R^{d-1}, there must be disjoint index sets S and T whose convex hulls intersect and whose corresponding intersections of A_i over S and -A_i over T also intersect. Fan's theorem is exactly the special case where X lies on the moment curve, so the alternating sign pattern is one order type among many. The same mechanism yields a continuous version replacing convex hulls by images of disjoint faces under a continuous map, a colorful version with d+1 separate coverings, and a parity strengthening of Fan's counting lemma. Applications include structural results for graph colorings, a new proof of the topological Hall theorem, and mass-partition statements that go beyond existence to prescribe which measures are pushed to which side of a hyperplane.","feed_headline":"Coverings of the sphere force Radon-pair intersections","feed_subtitle":"A structural upgrade of Borsuk–Ulam: not just a zero exists, but the sign pattern is fixed by any point configuration.","key_machinery":"The load-bearing object is Lemma 2.1: for X⊂R^{d-1}, write A^+={(x,1)} and B^-={(-x,-1)}; then conv A ∩ conv B ≠ ∅ if and only if 0∈conv(A^+∪B^-). This turns a convex-geometric Radon pair into a zero of an antipodally symmetric linear map on a triangulated sphere, so the Borsuk–Ulam theorem applies directly. For the continuous generalization, the analogous object is the antipodal map S^d→(Δ^{m-1})^{*2}_Δ built from a partition of unity subordinate to the projected covering; composing with the lifted continuous map converts an intersection of continuous images into an odd map whose zero is the desired point. Theorem 2.2 additionally uses degree: the relevant odd map has odd degree by Borsuk–U","core_discovery":"The paper establishes a transfer principle: Radon-type intersection combinatorics in Euclidean space constrain the intersection combinatorics of antipodal sphere coverings, and hence of every Borsuk–Ulam-type construction. The proof reduces the geometry to an odd map: lift each point x_i to (x_i,1) and each antipode to (-x_i,-1), label a sufficiently fine antipodal triangulation of the sphere with these points, invoke the Borsuk–Ulam theorem to get a zero, then let the triangulation mesh go to zero. A zero of the labeled map is equivalent to a Radon pair for X, so the limit produces disjoint faces S,T whose convex hulls intersect and whose corresponding covering sets intersect. The paper als","pith_inferences":["The Fan-complex discussion in Section 7 suggests that the full set of Radon pairs is not merely sufficient but essentially necessary: if a single minimal Radon pair is deleted, the pattern is no longer forced.","Because the main proof is constructive at positive mesh size and passes to the limit by compactness, the corresponding search problems plausibly inherit membership in PPA; the parity theorem may provide a second witness beyond the existence of an intersection.","The colorful setup of Theorem 2.6 appears ripe for iteration: combining it with other point configurations should yield colorful KKM, Komiya, and ham-sandwich consequences for every order type, not only the cyclic ones.","The product-sphere theorem is proved only for d a power of two; a natural test is whether this restriction is genuinely necessary for the structural conclusion or an artifact of the known obstruction."],"forward_implications":["Every order type of m points in R^{d-1} produces its own Fan-type covering theorem; Fan's alternating pattern is the special case of points on the moment curve, so the family of such results is much larger than the classical one.","For generic odd labelings of antipodal triangulations, not only does a zero-capturing facet exist, but the number of such facets is exactly twice an odd number, extending Fan's parity statement.","Proper colorings of graphs whose chromatic number is bounded via Borsuk–Ulam must contain forced rainbow structures: with enough colors, some complete bipartite subgraph has color sets on its two sides forming a Radon pair, yielding local-chromatic-number bounds.","The topological Hall theorem and hypergraph Hall-type results on systems of disjoint representatives follow as special cases of the rainbow-face version, giving structural statements about matchings beyond the classical dimension threshold.","For more than d masses in R^d, there exists a hyperplane that simultaneously places the measures indexed by one side of a Radon pair on opposite sides of itself; the (Z/2)^2 product-sphere version gives an analogous two-hyperplane chessboard partition when d is a power of two."],"supporting_citations":[{"why":"Fan's theorem, the classical result being generalized; supplies the alternating-sign-pattern special case.","marker":"[22]"},{"why":"Gale's evenness criterion, used to identify Fan's theorem as the moment-curve order type.","marker":"[29]"},{"why":"Radon's theorem and the convex-hull identity that converts Radon pairs into zeros of an odd map.","marker":"[56]"},{"why":"The Borsuk–Ulam theorem, the engine used to locate the zero in every proof.","marker":"[13]"},{"why":"Ramos's non-existence result for (Z/2)^2-equivariant maps, the foundation of the product-sphere generalization.","marker":"[57]"},{"why":"The authors' earlier colorful Borsuk–Ulam theorem, which Theorem 2.6 extends.","marker":"[27]"},{"why":"Aharoni–Haxell hypergraph Hall theorem, which the paper gives a new proof of and extends.","marker":"[3]"},{"why":"Meshulam's topological Hall theorem formulation, used as a comparison and special case.","marker":"[50]"},{"why":"van Kampen's non-embeddability theorem, recovered as an example of the continuous generalization.","marker":"[72]"}],"fun_headline_variants":["Borsuk–Ulam gets a structural upgrade via Radon pairs","Sphere coverings force Radon-pair intersections","Borsuk–Ulam's hidden structure laid bare by Radon pairs","From zero existence to forced geometry: Borsuk–Ulam generalized","New proof of topological Hall theorem from Borsuk–Ulam"],"cache_read_input_tokens":2688,"weakest_assumption_plain":"The load-bearing premise is that the covering sets can be approximated by arbitrarily fine antipodal triangulations while A_i and -A_i stay separated by a positive margin, so each vertex gets an unambiguous label and the zero of the limiting odd map survives; the product-sphere variant inherits the additional restriction that the underlying non-existence result is only proven when d is a power of two.","fun_headline_variants_meta":{"raw":{"variants":["Borsuk–Ulam gets a structural upgrade via Radon pairs","Sphere coverings force Radon-pair intersections","Borsuk–Ulam's hidden structure laid bare by Radon pairs","From zero existence to forced geometry: Borsuk–Ulam generalized","New proof of topological Hall theorem from Borsuk–Ulam"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000763,"raw_usage":{"total_tokens":3192,"prompt_tokens":682,"completion_tokens":2510,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":426,"completion_tokens_details":{"reasoning_tokens":2419}},"tokens_in":426,"tokens_out":2510,"duration_ms":21877,"temperature":1.0,"reasoning_tokens":2419,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-04T22:31:22.367316+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Set d=2 and take X={x1,x2,x3,x4} with x4 strictly inside the triangle x1x2x3, so the unique Radon pair is {4} against {1,2,3}. Try to construct closed sets A1,...,A4⊂S^2 with A_i∩(-A_i)=∅ and ∪(A_i∪(-A_i))=S^2 but with A_4∩(-A_1∩-A_2∩-A_3)=∅. Theorem 1.2 says no such covering exists; an explicit covering or a computational search showing one exists would refute it.","supporting_citations":[{"cited_title":"Math.56 (1952), no","cited_arxiv_id":null,"evidence_quote":"Fan's theorem, the classical result being generalized; supplies the alternating-sign-pattern special case."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Gale's evenness criterion, used to identify Fan's theorem as the moment-curve order type."},{"cited_title":"Ann.83(1921), no","cited_arxiv_id":null,"evidence_quote":"Radon's theorem and the convex-hull identity that converts Radon pairs into zeros of an odd map."},{"cited_title":"Math.20(1933), 177–190","cited_arxiv_id":null,"evidence_quote":"The Borsuk–Ulam theorem, the engine used to locate the zero in every proof."},{"cited_title":"Ramos,Equipartition of mass distributions by hyperplanes, Discrete Comput","cited_arxiv_id":null,"evidence_quote":"Ramos's non-existence result for (Z/2)^2-equivariant maps, the foundation of the product-sphere generalization."},{"cited_title":"(2025), no","cited_arxiv_id":null,"evidence_quote":"The authors' earlier colorful Borsuk–Ulam theorem, which Theorem 2.6 extends."},{"cited_title":"Graph Theory35(2000), no","cited_arxiv_id":null,"evidence_quote":"Aharoni–Haxell hypergraph Hall theorem, which the paper gives a new proof of and extends."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Meshulam's topological Hall theorem formulation, used as a comparison and special case."},{"cited_title":"van Kampen,Komplexe in euklidischen Räumen, Abh","cited_arxiv_id":null,"evidence_quote":"van Kampen's non-embeddability theorem, recovered as an example of the continuous generalization."}],"review_version":1}