{"id":"deda4910-a82f-496e-930c-100db026ec13","arxiv_id":"2411.19696","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"The Euler discriminant of families of hyperplane complements is the zero set of an explicit product of determinants indexed by connected square subgraphs.","lead":"The paper proves that the locus where a hyperplane arrangement changes topology, the Euler discriminant, is given by an explicit product of determinants. This gives a practical algebraic way to find singularities of integrals that appear in cosmology and particle physics.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The unqualified claim that the Euler discriminant is a hypersurface for hyperplane complements fails when χ* = 0, as the paper's own central-arrangement example shows; Theorem 4.1 needs its positivity hypothesis.","rationale":"The reader's verdict is CONDITIONAL and identifies positivity as the weakest assumption; I agree. The alternative concerns — the compressed proof of Theorem 5.4 and the conjectural multiplicity shortcut in §3.4 — are real but secondary: Theorem 5.4's equality would strengthen the D-module relation, and §3.4 affects computational practicality, not the truth of the combinatorial formula. The positivity issue, by contrast, changes the truth value of the paper's banner claim. Under the hypothesis, the proof of Theorem 4.1 is coherent: the matroid constancy on the complement of E_red is sound, the surjection (4.3)–(4.4) is standard, and the nonvanishing of ω_~I is the only place positivity is used. The counterexample the authors give is decisive: for central arrangements χ* = 0, ∇χ(Z) is empty, but E_red is not identically zero, so (4.2) fails. The paper is honest about this and explicitly calls the hypothesis fundamental, but the abstract and introduction do not carry the qualification. Since the fix is to restrict the statement rather than repair a proof step, the correct verdict remains CONDITIONAL; no change from the reader's assessment is needed. The concrete test would settle the scope issue by reproducing the counterexample and confirming the failure of (4.2).","tokens_in":28427,"tokens_out":7285,"duration_ms":65878,"concrete_test":"Test the failure mode on the paper's own central-arrangement example: take k = 2 and the arrangement of three lines through one point in P^2, parameterized as in the note after Theorem 4.1. Compute χ_z for all z in the subspace Z and verify it is identically zero, so ∇χ(Z) = ∅. Then compute E_red_χ(z) from (4.1) and verify it is not the zero function on Z, so the equality (4.2) fails. This directly checks whether the positivity hypothesis is necessary and whether the abstract's unqualified hypersurface claim must be restricted.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The most load-bearing concern is the scope of Theorem 4.1. The theorem's conclusion ∇χ(Z) = {E_red_χ = 0} is proved only under the hypothesis χ_z > 0 for some z ∈ Z, and the proof uses that hypothesis at the last step: the zero set V of the one-form (4.5) has cardinality χ* > 0, so the evaluation map O*/I → C^V is an isomorphism and the class ω_~I is shown nonvanishing by evaluating det(z*_{I,J})/(h_{i1}⋯h_{ik}) at points of V. If χ* = 0, then V is empty and this argument collapses. The paper itself notes that central arrangements satisfy χ_z = 0 for all z while E_red_χ is not identically zero; in that case ∇χ(Z) is empty and (4.2) is false. Consequently the abstract's unqualified assertion that the Euler discriminant is a hypersurface for complements of hyperplanes is not correct in general: it is true only for families with positive signed Euler characteristic. This is not a defect in the proof under the hypothesis, but it is a genuine restriction on the central claim as stated, and it is not merely a technicality since central arrangements are a natural class.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies the Euler discriminant of families of very affine varieties obtained as complements of hyperplanes, defined as the locus where the signed Euler characteristic drops from its maximal value. The two main theorems are a product-of-determinants formula for the principal A-determinant of a sparse hyperplane arrangement in terms of the face structure of an associated edge polytope (Theorem 3.9), and a determinantal characterization of the Euler discriminant when the coefficient parameters are constrained to an arbitrary smooth subvariety (Theorem 4.1). The paper also compares the Euler discriminant with the singular locus of the Gauss-Manin D-module attached to the Euler integral (Theorems 5.1–5.4), with the final equality restricted to the arrangement case. An appendix develops a connection to cosmological correlators and computes explicit Euler discriminants for two-site and three-site chain graphs.","tokens_in":28615,"tokens_out":20489,"duration_ms":175638,"significance":"If the results stand, they give the first general explicit formulas for Euler discriminants of hyperplane complements in a broad range of coefficient families, with direct applications to Euler integrals, Feynman integrals, and cosmological correlators. Theorem 3.9 is a genuinely combinatorial closed formula for the principal A-determinant in the sparse case, with a companion implementation, and Theorem 4.1 reduces the computation to products of square subdeterminants. The worked examples, including the n=6,k=2 graph and the cosmological chains, are concrete and reproducible, and the degree checks against the general degree formula provide useful confirmation. The main caveat, properly acknowledged in the body, is that Theorem 4.1 requires the positivity hypothesis χ_z>0, without which the Euler discriminant need not be a hypersurface.","major_comments":[{"comment":"The unqualified statement that the Euler discriminant is a hypersurface is not correct in the full generality claimed by the abstract and the introduction. Theorem 4.1 is proved only under the hypothesis that χ_z > 0 for some z ∈ Z, and the note after the theorem explicitly records a family with χ_z = 0 for all z for which E_red^χ is not identically zero; in that case ∇χ(Z) is empty, so (4.2) fails. The abstract and the sentence 'A clear consequence of Theorems 3.9 and 4.1 is that the Euler discriminant corresponds to locus where the matroid changes' must therefore be qualified to families with positive signed Euler characteristic, and the zero case should be described in the introduction rather than only in an afterthought.","section":"Abstract and p.2, Theorem 4.1 (Eq. 4.2)"},{"comment":"The statement 'For generic (s, ν), one has Sing(M) = ∇χ(Z)' omits the hypothesis χ_z > 0. The proof invokes 'Since ∇χ(Z) is purely one codimensional', which is a consequence of Theorem 4.1 only under that hypothesis; in the central-arrangement case ∇χ(Z) can be empty while Sing(M) is generally nonempty. The theorem statement should either include the positivity hypothesis or explicitly discuss the zero case and state what remains true in that case.","section":"Theorem 5.4, Section 5.3"}],"minor_comments":[{"comment":"The counterexample 'k=2, n=1, and z = [0 z11 0]^T' is not readable as written: it uses dimensions inconsistent with the earlier convention n−k (which would be negative) and an entry z_{11} that does not match the indexing z_{ij} with j ≥ k+1. Please replace it with a correctly specified central arrangement, e.g., a pencil of lines through a point in P^2, and verify the claimed χ_z=0 and non-vanishing of E_red^χ.","section":"Note after Theorem 4.1, p.19"},{"comment":"The sentence 'Using Theorem 4.1 we compute the factors of the principal A-determinant E_{A_G}(z_G)' should refer to Theorem 3.9, which is the theorem giving the product formula in the sparse case.","section":"Example 3.10, p.13"},{"comment":"The assertion that the complement of the right-hand side has non-empty intersection with the locus where the Euler characteristic attains its maximum is not justified in the text. Please add a justification, for example by upper semicontinuity of z ↦ χ_z or by a reference establishing that the generic (maximally uniform) matroid stratum attains the maximal beta invariant.","section":"Proof of Theorem 4.1, p.18"},{"comment":"Condition (*) is defined only for graphs with |V1|=|V2|, but Theorem 3.9 applies it to subgraphs G_{I∪J} after imposing |I|=|J|. Please state explicitly that (*) is applied to the subgraph with left vertex set I and right vertex set J.","section":"Section 3.3, condition (*)"},{"comment":"There are several typos and minor presentation issues: 'sigend' (p.7), 'showes' (p.3), 'deteailed' (p.27), 'M¨ unch' (p.31), and the 'factors in cerulean' in Example A.7 (p.30) are not visible in a monochrome rendering and should be labeled explicitly.","section":"Various"}],"recommendation":"minor_revision","confidential_remarks":"The mathematical core of the paper is sound under the stated hypotheses, and the companion implementation and worked examples are a strength. The main issue is the overstatement in the abstract and in Theorem 5.4 about the generality of the hypersurface conclusion; this is local and fixable by qualification. I see no reason to doubt novelty or provenance. The paper is a good fit for the journal's scope."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Two things to know about this paper. The core results are real: Theorem 3.9 gives an explicit product formula for the principal A-determinant of a sparse hyperplane arrangement, including multiplicities, and Theorem 4.1 gives a determinantal description of the Euler discriminant for coefficient spaces that are arbitrary subvarieties, under a stated positivity assumption. The paper is honest about the assumption: it flags the central-arrangement case where χz=0 and the formula fails. The abstract, however, overstates the case by saying the Euler discriminant is a hypersurface for complements of hyperplanes, unqualified; that is only true when the signed Euler characteristic is positive somewhere. This should be fixed, but it does not undermine the main theorems.\n\nCredit where it is earned: the combinatorial core of Theorem 3.9 is proven from face decompositions of edge polytopes and matroid arguments, and checked against degree identities and worked examples. The multiplicity computation via subdiagram volumes is a genuine step beyond the normal-fan description in Galashin–Nenashev–Postnikov. The extension to arbitrary subvarieties in Theorem 4.1 is new and useful for applications like Feynman and cosmological integrals, and the appendix gives concrete worked examples with code available.\n\nSoft spots: The proof of Theorem 5.4, identifying the D-module singular locus with the Euler discriminant in the arrangement case, compresses the key direct-image and transversality step into a citation to a standard lemma without the needed verification in this setting. It is plausible but under-argued; a referee should ask for a fuller proof. Section 3.4's shortcut for computing multiplicities via Hilbert polynomials of associated graded rings is only conjectural beyond the examples; the authors say 'seems to hold in general.' That is fine as a computational heuristic, but it should not be presented as a proven method. The positivity hypothesis is not a flaw in the proof, but it is a genuine restriction, and the abstract's wording currently hides it.\n\nBottom line: this deserves peer review. A careful referee can sort out the abstract and the D-module proof. The main combinatorial content is solid and useful. I would bring it to a reading group.","headline":"Solid combinatorial results with an overbroad abstract: the Euler discriminant is a hypersurface only under the paper's own positivity hypothesis.","tokens_in":29213,"tokens_out":2564,"would_cite":true,"duration_ms":20909,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["32S22","14M25","14F10"],"pacs":[],"model":"deepseek-v4-flash","headline":"The paper proves that for families of complements of hyperplanes, the Euler discriminant—the locus where the signed Euler characteristic drops—is a hypersurface, and that under a positivity assumption it is exactly the zero locus of the…","keywords":["Euler discriminant","hyperplane arrangements","very affine varieties","principal A-determinant","Euler integrals","edge polytopes","bipartite graphs","D-modules"],"falsifier":"The paper's own central-arrangement example ($k=2$, $n=1$, coefficient matrix $z=[0,z_{11},0]^T$) has $\\chi_z=0$ for every $z$ while the reduced discriminant is a non-zero polynomial, so $\\nabla_\\chi(Z)$ is not its zero locus; reproducing that computation confirms the positivity hypothesis is necessary. Conversely, any family with $\\chi_z>0$ somewhere where the two sets differ would refute Theorem 4.1.","tokens_in":28166,"feed_emoji":"📐","tokens_out":13372,"duration_ms":97202,"temperature":0.7,"pith_summary":"Variation of the signed Euler characteristic of hyperplane complements controls the singularity structure of Euler integrals, which appear in Feynman and cosmological correlation computations. This paper shows that the locus where that Euler characteristic drops—the Euler discriminant—is a hypersurface in the space of coefficients and identifies its defining equation. In the sparse-coefficient case the equation is a product of determinants of square submatrices, with multiplicities given by combinatorial subdiagram volumes of an associated bipartite graph. In the general case of coefficients constrained to a smooth subvariety, the same zero-locus description holds provided the signed Euler characteristic is positive somewhere. A consequence is that the Euler discriminant is computable directly from matroid data and coincides with the singular locus of the associated Gauss–Manin connection in the hyperplane case.","feed_headline":"Euler discriminant of hyperplane complements is a hypersurface","feed_subtitle":"When coefficients vary generically or in a subspace, the drop locus is the zero set of a product of determinants.","key_machinery":"The carrying object is the reduced discriminant, the product of all square subdeterminants $\\det(z_{I,J})$ of the coefficient matrix that are not identically zero on the chosen coefficient space. In the sparse case the same object arises from the edge polytope of a bipartite graph G: the factors are $\\det(z_{I,J})$ for equal-size subsets I of the left vertices and J of the right vertices whose induced subgraph is connected and satisfies condition (∗), a Hall-type condition that makes the determinant irreducible; the exponents are subdiagram volumes computed from a contracted graph $G/H$ via non-homogeneous toric ideals. The proof that the zero locus of this product is exactly the drop locus runs through the Orlik–Solomon algebra: the matroid of the arrangement is constant on the complement of the zero locus, and a vanishing minor produces a nonzero class in the top cohomology of the generic matroid that maps to zero in the specialized matroid, so the dimension—the signed Euler characteristic—must drop.","core_discovery":"The central discovery is that the Euler discriminant of a family of hyperplane complements is governed by square determinants of the coefficient matrix. Theorem 4.1 states that if the signed Euler characteristic is positive for at least one point of a smooth coefficient subvariety Z, then the Euler discriminant equals the zero locus in Z of the reduced discriminant—the product of all square subdeterminants that do not vanish identically on Z. The proof shows that away from this zero locus the matroid of the arrangement is constant, so the signed Euler characteristic attains its maximal value, while on the zero locus a vanishing minor produces a nonzero cohomology class in the generic Orlik–Solomon algebra that dies in the specialized one, forcing the Euler characteristic to drop. In the sparse coordinate-subspace case, Theorem 3.9 sharpens the product to run only over square submatrices whose induced bipartite subgraph is connected and satisfies a Hall-type condition, with multiplicities equal to subdiagram volumes. The paper additionally proves that for hyperplane complements the Euler discriminant coincides exactly with the singular locus of the D-module underlying the Euler integral.","pith_inferences":["Since the Euler discriminant is the locus where the matroid changes, the matroid stratification of the coefficient space refines the Euler stratification; testing whether the two agree on small examples would show how much information the Euler discriminant retains.","In the cosmological integrals of the appendix, some factors of the Euler discriminant (like $Y_{12}=0$ in the two-site chain) are discarded as artifacts of normalization; a monodromy analysis of the twisted cycle could convert this heuristic into a principled rule for selecting physical singularities.","The positivity failure in central arrangements suggests that a limiting form of the determinantal formula might describe the Euler discriminant with adjusted multiplicities, but the paper's example indicates that no immediate product-of-minors formula survives without modification."],"forward_implications":["Where the positivity hypothesis holds, the Euler discriminant of a family of hyperplane complements is a hypersurface cut out by the reduced discriminant, so it can be computed by linear algebra on the coefficient matrix without evaluating any Euler characteristic.","The Euler discriminant is the locus where the matroid of the hyperplane arrangement changes, making the drop in Euler characteristic a matroid invariant.","For sparse arrangements, the principal A-determinant has an explicit product formula indexed by balanced connected bipartite subgraphs satisfying condition (∗), giving both its Newton polytope and the multiplicities of its components.","The singular locus of the D-module annihilating an Euler integral coincides with the Euler discriminant in codimension one, and exactly in the hyperplane-complement case, so the singularity structure of such integrals is readable from the determinants."],"supporting_citations":[{"why":"Defines the principal A-determinant and A-discriminant, the objects whose zero loci the paper relates to the Euler discriminant in the full and sparse coefficient cases.","marker":"[19]"},{"why":"Introduced the Euler discriminant for Feynman integrals and supplies the computational tool used to obtain component multiplicities in the examples.","marker":"[16]"},{"why":"First defined the Euler discriminant as a divisor with multiplicities given by Euler-characteristic drops, and the paper invokes its Theorem 3.3 for multiplicities.","marker":"[14]"},{"why":"Describes the facet structure of edge polytopes of bipartite graphs, the key combinatorial input for the sparse-case product formula in Theorem 3.9.","marker":"[33]"},{"why":"Hall's matching theorem characterizes when a bipartite graph has a perfect matching, which underlies condition (∗) and the irreducibility of $\\det(z_{I,J})$.","marker":"[32]"},{"why":"Identifies the dimension of hyperplane-complement local system cohomology with the signed Euler characteristic, used in Theorem 4.1 to measure the drop.","marker":"[13]"},{"why":"Provides the result that the zero set of a generic logarithmic one-form has cardinality equal to the signed Euler characteristic, which is the source of the positivity hypothesis.","marker":"[23]"},{"why":"Gives the beta-invariant interpretation of the signed Euler characteristic and its equality with the number of bounded regions for real arrangements, framing the discriminant as the locus where bounded chambers shrink.","marker":"[35]"}],"fun_headline_variants":["Euler discriminant: zero locus of square determinants","Hyperplane complements: Euler discriminant is a hypersurface","Where Euler characteristic drops: product of square determinants"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the signed Euler characteristic is positive for at least one choice of coefficients in the family; the paper itself notes that for central arrangements the signed Euler characteristic is zero everywhere and the determinantal description no longer holds.","fun_headline_variants_meta":{"raw":{"variants":["Euler discriminant: zero locus of square determinants","Hyperplane complements: Euler discriminant is a hypersurface","Where Euler characteristic drops: product of square determinants"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001724,"raw_usage":{"total_tokens":6782,"prompt_tokens":871,"completion_tokens":5911,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":487,"completion_tokens_details":{"reasoning_tokens":5863}},"tokens_in":487,"tokens_out":5911,"duration_ms":33472,"temperature":1.0,"reasoning_tokens":5863,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T05:56:41.774880+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"The paper's own central-arrangement example ($k=2$, $n=1$, coefficient matrix $z=[0,z_{11},0]^T$) has $\\chi_z=0$ for every $z$ while the reduced discriminant is a non-zero polynomial, so $\\nabla_\\chi(Z)$ is not its zero locus; reproducing that computation confirms the positivity hypothesis is necessary. Conversely, any family with $\\chi_z>0$ somewhere where the two sets differ would refute Theorem 4.1.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Defines the principal A-determinant and A-discriminant, the objects whose zero loci the paper relates to the Euler discriminant in the full and sparse coefficient cases."},{"cited_title":"Fevola, S","cited_arxiv_id":null,"evidence_quote":"Introduced the Euler discriminant for Feynman integrals and supplies the computational tool used to obtain component multiplicities in the examples."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"First defined the Euler discriminant as a divisor with multiplicities given by Euler-characteristic drops, and the paper invokes its Theorem 3.3 for multiplicities."},{"cited_title":"Ohsugi and T","cited_arxiv_id":null,"evidence_quote":"Describes the facet structure of edge polytopes of bipartite graphs, the key combinatorial input for the sparse-case product formula in Theorem 3.9."},{"cited_title":"Murthy and J","cited_arxiv_id":null,"evidence_quote":"Hall's matching theorem characterizes when a bipartite graph has a perfect matching, which underlies condition (∗) and the irreducibility of $\\det(z_{I,J})$."},{"cited_title":"Esnault, V","cited_arxiv_id":null,"evidence_quote":"Identifies the dimension of hyperplane-complement local system cohomology with the signed Euler characteristic, used in Theorem 4.1 to measure the drop."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the result that the zero set of a generic logarithmic one-form has cardinality equal to the signed Euler characteristic, which is the source of the positivity hypothesis."},{"cited_title":"Orlik and H","cited_arxiv_id":null,"evidence_quote":"Gives the beta-invariant interpretation of the signed Euler characteristic and its equality with the number of bounded regions for real arrangements, framing the discriminant as the locus where bounded chambers shrink."}],"review_version":1}