{"id":"97b65ce0-0db0-44b3-b927-b834838497c0","arxiv_id":"1908.08573","paper_version":1,"verdict":"REJECT","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"high","formal_verification":"none","parameter_count":0,"one_line_summary":"Calabi-Yau manifolds with a non-Hodge-degenerate mirror dual are Kobayashi non-hyperbolic, assuming the BCOV relation between genus-one Gromov-Witten invariants and Ray-Singer torsion.","lead":"This paper argues that any Calabi-Yau manifold whose mirror dual exists and has a non-constant BCOV torsion over its moduli space must contain rational or elliptic curves, hence is Kobayashi non-hyperbolic. The argument is a short application of mirror symmetry and the BCOV holomorphic anomaly equation, but it rests on physics conjectures rather than proven mathematics.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Theorem 3.1 rests on an unproven exact identification between F1 and BCOV torsion; even granting it, the step from ∂∂ log T ≠ 0 to a nonzero genus-one q-series needs a normalization that the paper never supplies.","rationale":"The paper aims to prove a mathematical theorem in differential geometry, but the proof's decisive sentence is an appeal to [BCOV], a physics derivation, followed by an unproven implication. The reader's rejection is justified. I do not see an internal contradiction in the Hodge-degeneracy computations (Lemmas 2.8, Remarks 2.6-2.7); those parts are consistent. The missing piece is the bridge from BCOV torsion anomaly to nonzero GW invariants. This is not merely a matter of consensus: the exact normalization of F1 is part of the claim, and without it the q-series could vanish while ∂∂ log T is nonzero. A concrete re-derivation of this step would settle whether the theorem can be made rigorous. The reader identifies the BCOV relation as the weak assumption; I agree but sharpen it to the unsupplied normalization and the missing inference from ∂∂ log T to the GW q-series. Thus the recommended verdict is unchanged.","tokens_in":5356,"tokens_out":13823,"duration_ms":153152,"concrete_test":"Starting from (2.2) and the definition of F1 as the generating function of genus-one GW invariants, attempt to prove the sentence 'for which it is enough that ∂∂ log T ≠ 0' in the proof of Theorem 3.1: write F1 - (1/24)∫ k c2 as log T + h + \\bar h and determine whether ∂∂ log T ≠ 0 at a point forces the large-radius q-series of F1 - (1/24)∫ k c2 to have a nonzero coefficient. If the derivation requires an additional normalization of the holomorphic ambiguity h (e.g., h = 0 at the large-radius point), then the theorem as stated is incomplete; if the derivation goes through without extra assumptions, the concern is refuted.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is carried by one sentence in the proof of Theorem 3.1: 'By [BCOV], F1 corresponds to log T ... for which it is enough that ∂∂ log T ≠ 0.' This packages two unproven assumptions. First, [BCOV] is a physics derivation, not a mathematical theorem, and in physics the topological string amplitude F1 is defined only up to addition of a holomorphic function of the moduli, so 'corresponds' is not an equality with a specified normalization. Second, even if a normalization is fixed, the text never proves that ∂∂ log T ≠ 0 forces F1 minus the constant-map contribution to have a nonzero positive-degree q-coefficient. Nonzero ∂∂ of a function does not by itself prevent the function from vanishing at the relevant point; what is needed is a statement about the large-radius q-expansion of the genus-one GW invariants and the behavior of the holomorphic ambiguity under the mirror map. Equations (2.2)-(2.5) merely translate ∂∂ log T into Chern classes of Hodge bundles; they do not connect it to GW invariants. Thus, if the BCOV identification fails, or if the unspecified holomorphic ambiguity cancels the positive-degree terms, the conclusion does not follow. The paper is a conditional research announcement depending on a physics conjecture, not an established mathematical theorem.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper claims to establish Kobayashi non-hyperbolicity of a Calabi-Yau manifold X provided that a mirror dual ˇX exists and is not 'Hodge degenerate.' The argument proceeds through the BCOV relation between the genus-one topological string partition function F1 of X and the logarithm of the holomorphic Ray-Singer torsion T of ˇX, and asserts that if ∂∂ log T ≠ 0 then F1 differs from its constant-map contribution, which in turn yields nontrivial rational or elliptic curves and hence an entire curve. The paper also derives a characterization of Hodge degeneracy for Calabi-Yau threefolds in terms of the Ricci form of the Weil-Petersson metric. The theorems are conditional on the existence of a mirror and on the BCOV identification, and the logical step from ∂∂ log T ≠ 0 to nonvanishing Gromov-Witten invariants is not derived.","tokens_in":5647,"tokens_out":8859,"duration_ms":82196,"significance":"If the argument were made rigorous, the result would be a major contribution to the Kobayashi conjecture for Calabi-Yau manifolds, covering a very wide class of examples and forging a new link between analytic torsion and hyperbolic geometry. The paper's clear formulation of a sufficient condition in terms of Hodge bundles is a useful conceptual step, and the explicit treatment of the quintic and of K3 surfaces helps illustrate the mechanism. However, the central implications rest on unproven physics conjectures and on an unproven inference about q-expansions. The theorems are best read as conditional statements, and the paper does not provide the missing mathematical arguments that would substantiate the abstract's claim of 'proving' Kobayashi non-hyperbolicity.","major_comments":[{"comment":"The inference 'for which it is enough that ∂∂ log T ≠ 0' is not justified. A nonzero ∂∂ of a function does not by itself imply that the function has a nonzero positive-degree q-expansion; for example, any holomorphic function has vanishing ∂∂, and the constant-map subtraction is a holomorphic expression in the Kähler class. The proof needs an explicit argument showing that ∂∂ log T ≠ 0 forces at least one positive-degree genus-one invariant N_{1,d} (or a genus-zero invariant) to be nonzero after fixing the holomorphic ambiguity and the mirror map. Equations (2.2)–(2.5) relate ∂∂ log T to Chern classes of Hodge bundles and the Weil-Petersson form, but they do not connect it to the q-expansion coefficients of F1. Without such an argument, the implication is not a logical consequence of the stated assumptions.","section":"Section 3, proof of Theorem 3.1"},{"comment":"The identification F1 = log T is cited from [BCOV], a physics paper, and is used as an exact equality. No normalization of F1, of the mirror map, or of the holomorphic ambiguity is specified, and no proof or precise statement of the equality is supplied. Consequently Theorems 3.1 and 3.2 are conditional on the BCOV conjecture and on the existence of a mirror dual, not established mathematical results. The abstract's phrasing 'we establish ... therefore proving' overstates the status of the argument. If the authors intend to present conditional results, that should be stated explicitly and the conditional derivation should be rigorous; as written, the central claim is not a theorem in the mathematical sense.","section":"Section 2, Theorem 2.3 and Section 3, proof of Theorem 3.1"}],"minor_comments":[{"comment":"The notation ∆′_{p,q} and '∂-Laplace operator' should be clarified: presumably the intended operator is the ∂̄-Laplacian on (p,q)-forms, and the regularized determinant and the product over (p,q) should be precisely defined.","section":"Section 2, Definition 2.2"},{"comment":"The displayed non-Hodge-degeneracy condition 'Ric(ω_WP) ≠ -(m+3 - χ(X)/12)ω_WP' is correct only when applied to the mirror ˇX, using χ(ˇX) = -χ(X). The text should state this explicitly to avoid the appearance of a sign error relative to Lemma 2.8.","section":"Section 3, Theorem 3.1"},{"comment":"The equivalence between 'Kobayashi non-hyperbolic' and the existence of an entire curve is used throughout but never stated; the authors should note that they are using Brody's theorem for compact complex manifolds.","section":"Introduction and abstract"},{"comment":"There are several typographical and stylistic issues: for example, 'invarian ts' in the Introduction, and the reference to [Z2] and [FLY] in Remark 2.11 should be more precise about what exactly is proven and by whom.","section":"Throughout"}],"recommendation":"reject","confidential_remarks":"The paper relies heavily on the BCOV paper, which includes a coauthor of this manuscript, but I do not regard this as circular; it is a standard physics conjecture. The main problem is that the mathematical argument is incomplete and, in part, logically unjustified. This is a very short research announcement; for a mathematics journal, the theorems should either be explicitly conditional with a complete derivation, or be replaced by a detailed proof. As it stands, the paper is better suited to a physics or mathematical physics venue. I recommend rejection, though I would encourage the authors to develop the missing argument in a future version."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear [Colleague],\n\nThe one-sentence take: this paper is a short, honest research announcement that turns a known physics identification into a geometric sufficient condition for Kobayashi non-hyperbolicity. It is not a complete proof, but it deserves peer review if framed as conditional.\n\nWhat is actually new: the observation that if the mirror of a Calabi-Yau X is not 'Hodge degenerate' in the sense of equation (2.5), then the BCOV torsion log T has nonzero ∂∂, and, assuming the mirror relation F1 = log T, this should force non-trivial genus-one or genus-zero Gromov–Witten invariants, hence rational or elliptic curves on X. That is a genuinely new bridge between the Hodge-bundle geometry of the mirror moduli space and curve-counting. The paper is also honest: it repeatedly states the mirror assumption and notes that K3s and tori are Hodge degenerate but nonetheless non-hyperbolic.\n\nThe problems are where you would expect. The BCOV relation is a physics derivation, not a theorem, and F1 is only defined up to a holomorphic function of the moduli. The step 'for which it is enough that ∂∂ log T ≠ 0' is asserted, not derived. Even granting the identification, one needs a normalization: ∂∂ log T ≠ 0 says the torsion is not pluriharmonic on the moduli space, but it does not directly say anything about the q-expansion coefficients of F1 minus the constant-map term. The paper should either supply that argument or explicitly state the result as a conjecture with the necessary normalization. Also, they produce no new examples; the quintic case is already known via Zinger.\n\nThe citation pattern is fine. [BCOV] is the obvious source for the torsion relation and Vafa being a coauthor there is not a problem in itself.\n\nBottom line: this is a conditional theorem at best, but the condition is well-defined and the connection is worth having on record. If it goes to review, the editor should ask for a precise statement of the BCOV identification and a proof of the q-expansion step, or else for the paper to be titled as a conjecture. I would accept it for peer review and send the authors those questions rather than desk-reject.","headline":"A short, conditional research announcement that deserves peer review if the BCOV normalization and q-expansion step are made explicit.","tokens_in":6160,"tokens_out":1837,"would_cite":true,"duration_ms":18213,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["32Q45","14J32","14N35","32G20"],"pacs":[],"model":"deepseek-v4-flash","headline":"Using mirror symmetry, the paper establishes that any Calabi-Yau manifold whose mirror dual exists and is not Hodge degenerate contains rational or elliptic curves, and hence is Kobayashi non-hyperbolic.","keywords":["Calabi-Yau manifold","Kobayashi non-hyperbolicity","mirror symmetry","Gromov-Witten invariant","Ray-Singer torsion","Hodge bundle","entire curve","topological string"],"falsifier":"Compute both sides of the identity $F_1 = \\log T$ for a mirror pair outside the quintic, for example a Calabi-Yau complete intersection in a toric variety; if the nonconstant part of the genus-one amplitude does not match the mirror's torsion variation, the central mechanism is false. Equivalently, exhibit a Calabi-Yau $X$ whose mirror has $\\partial\\bar\\partial\\log T\\neq 0$ but whose genus-zero and genus-one Gromov-Witten invariants all vanish, so no rational or elliptic curve is present.","tokens_in":5169,"feed_emoji":"🪞","tokens_out":15228,"duration_ms":129972,"temperature":0.7,"pith_summary":"This paper attacks the weak form of Kobayashi's conjecture, which says that no compact Calabi-Yau manifold is Kobayashi hyperbolic: every one should admit a nonconstant holomorphic map from the complex plane. The authors prove this for any Calabi-Yau manifold $X$ whose mirror dual $\\check X$ exists and is not Hodge degenerate, a curvature condition on the Hodge bundles over the mirror moduli space. The proof runs through mirror symmetry: the genus-one topological string partition function on $X$ is identified with the logarithm of the holomorphic Ray-Singer torsion of $\\check X$, and whenever that torsion varies nontrivially, nonvanishing Gromov-Witten invariants force the existence of rational or elliptic curves on $X$. Since no simply connected higher-dimensional Calabi-Yau manifold with maximal holonomy is known to be Hodge degenerate, the result in practice covers all known examples.","feed_headline":"Mirror symmetry forces curves onto Calabi-Yau manifolds","feed_subtitle":"A mirror-symmetry proof supports Kobayashi's conjecture for Calabi-Yau manifolds with non-degenerate mirrors.","key_machinery":"The engine is the BCOV relation between the genus-one topological string amplitude on $X$ and the holomorphic Ray-Singer torsion of its mirror, together with the Hodge-degeneracy condition it induces. On the mirror moduli space, $\\partial\\bar\\partial\\log T$ is expressed through the Hodge metric forms $\\omega_{H^i}$ and the Weil-Petersson form $\\omega_{WP}$, and the paper calls $\\check X$ Hodge degenerate when this quantity vanishes, equivalently when a certain Chern-class identity (equation 2.5) holds. In three dimensions that degeneracy condition is equivalent to the Ricci curvature of the Weil-Petersson metric being proportional to the metric with a specific coefficient. The contrapositive drives the proof: a non-Hodge-degenerate mirror gives $\\partial\\bar\\partial\\log T\\neq 0$, hence a nonconstant genus-one amplitude on $X$, hence a nonzero genus-one or genus-zero Gromov-Witten invariant, hence a rational or elliptic curve.","core_discovery":"For a compact Calabi-Yau $n$-fold $X$ with $n>2$ whose mirror $\\check X$ exists and is not Hodge degenerate, $X$ is Kobayashi non-hyperbolic. Hodge degeneracy is defined by $\\partial\\bar\\partial\\log T = 0$ for the mirror's holomorphic Ray-Singer torsion $T$; if the mirror is not Hodge degenerate, this quantity is nonzero. Through the BCOV identity the genus-one partition function on $X$ is $F_1 = \\log T$, so the nonconstant part of $F_1$ is nonzero, forcing some genus-one or genus-zero Gromov-Witten invariant to be nonzero. Such an invariant counts (possibly singular) rational or elliptic curves in $X$, and any such curve is an entire curve, so $X$ is Kobayashi non-hyperbolic. For threefolds the condition reduces to a concrete non-proportionality statement about the Ricci curvature of the Weil-Petersson metric.","pith_inferences":["The same mechanism could yield degree-dependent lower bounds on counts of rational or elliptic curves, since the nonvanishing Gromov-Witten invariant carries a curve degree, a quantitative refinement not pursued in the paper.","If the BCOV identity were proved or verified for more mirror pairs, the theorem would make Kobayashi non-hyperbolicity a computable check on the mirror moduli space rather than a geometric existence question.","The known Hodge-degenerate cases (K3 surfaces and complex tori) are themselves non-hyperbolic, suggesting the non-degeneracy condition is sufficient but may not mark the true boundary of the conjecture."],"forward_implications":["Every Calabi-Yau threefold with a mirror whose Weil-Petersson Ricci curvature is not proportional to the metric is Kobayashi non-hyperbolic.","Non-hyperbolicity is achieved by possibly singular rational or elliptic curves, so smooth curves are not required, which complements earlier results for Calabi-Yau threefolds with large Picard number.","Because no simply connected maximal-holonomy Calabi-Yau of dimension greater than two is known to be Hodge degenerate, the theorem brings the weak Kobayashi conjecture within reach of essentially all known examples, conditional on existence of a mirror.","For quintic threefolds the mechanism is already explicit: known computations of reduced genus-one Gromov-Witten invariants and of the mirror BCOV torsion force non-hyperbolicity."],"supporting_citations":[{"why":"Supplies the core identity $F_1=\\log T$ and the holomorphic anomaly equation (2.2) from which the proof starts.","marker":"[BCOV]"},{"why":"Gives the expression of $\\partial\\bar\\partial\\log T$ in terms of Hodge metric and Weil-Petersson forms, which defines Hodge degeneracy.","marker":"[FL]"},{"why":"Computes the BCOV torsion of the quintic mirror, providing the concrete example where the torsion variation can be checked.","marker":"[FLY]"},{"why":"Relates ordinary and reduced genus-one Gromov-Witten invariants, so nonvanishing invariants imply genus-one or genus-zero curves.","marker":"[Z1]"},{"why":"Computes the genus-one generating function for quintic threefolds and matches it to the mirror BCOV torsion, corroborating the mechanism.","marker":"[Z2]"},{"why":"Formulates Kobayashi hyperbolicity and the conjecture that Calabi-Yau manifolds are non-hyperbolic, which the paper addresses.","marker":"[K]"}],"fun_headline_variants":["Mirror symmetry reveals curves on Calabi-Yau manifolds","Calabi-Yau manifolds lose hyperbolicity via mirror symmetry","Kobayashi non-hyperbolic Calabi-Yau from mirror duals","Mirror duals force curves onto Calabi-Yau manifolds"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is the physics identity that the genus-one topological string partition function on $X$ equals the logarithm of the holomorphic Ray-Singer torsion of the mirror $\\check X$, an equality not proved as a theorem for general Calabi-Yau manifolds; if it fails in a given case, the deduction of curves from nonzero torsion variation no longer goes through.","fun_headline_variants_meta":{"raw":{"variants":["Mirror symmetry reveals curves on Calabi-Yau manifolds","Calabi-Yau manifolds lose hyperbolicity via mirror symmetry","Kobayashi non-hyperbolic Calabi-Yau from mirror duals","Mirror duals force curves onto Calabi-Yau manifolds"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001921,"raw_usage":{"total_tokens":7464,"prompt_tokens":831,"completion_tokens":6633,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":447,"completion_tokens_details":{"reasoning_tokens":6557}},"tokens_in":447,"tokens_out":6633,"duration_ms":46572,"temperature":1.0,"reasoning_tokens":6557,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T11:36:24.404252+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute both sides of the identity $F_1 = \\log T$ for a mirror pair outside the quintic, for example a Calabi-Yau complete intersection in a toric variety; if the nonconstant part of the genus-one amplitude does not match the mirror's torsion variation, the central mechanism is false. Equivalently, exhibit a Calabi-Yau $X$ whose mirror has $\\partial\\bar\\partial\\log T\\neq 0$ but whose genus-zero and genus-one Gromov-Witten invariants all vanish, so no rational or elliptic curve is present.","supporting_citations":[],"review_version":1}