{"id":"d2e01241-e98c-4f06-895f-140842558196","arxiv_id":"1909.01842","paper_version":3,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"For k>1, the noncompact Calabi-Yau threefold W_k admits infinitely many pairwise non-isomorphic complex deformations, and some of these deformations preserve nontrivial moduli spaces of rank-2 vector bundles.","lead":"This paper constructs noncompact Calabi-Yau threefolds with infinitely many non-isomorphic deformations and shows that some deformations still carry positive-dimensional moduli spaces of vector bundles. Readers interested in deformation theory or string theory on local Calabi-Yau geometries will find explicit families and computations.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"No significant objection identified: the h^1 computation in Thm 1.12 and the deduction of Cor 1.29 survive a detailed check.","rationale":"The central claim is Corollary 1.29: for k > 1, W_k has infinitely many distinct deformations. The invariant behind Theorem 1.13 is h^1(W2(y), TW2(y)) = y-1, computed in Theorem 1.12. I checked this computation in detail. The potentially fragile assertion is that no monomial z^{-1}u1^0u2^s with s < y-1 appears on the right-hand side of the coboundary equation. This assertion is correct: α terms have nonnegative z-powers; β terms contribute minimum u2-degrees y (from β1 u2^y and β2), and y-1 (from β3), so no cancellation can create smaller exponents. Lemma 1.9's reduction is also valid because the tangent transition raises the z-power of second-row terms by 2, leaving only the l=-1, i=0 part as a negative power of ξ. The paper's proof of Cor 1.29 is very terse, but the composition can be justified explicitly by pulling W2(y) back through the bundle isomorphism of Theorem 1.28 and introducing a parameter t. The resulting transition functions v1 = z^k u1 + t z^2 u2 + t^2 z u1^y, v2 = z^{-k+2}u2 + t z^{-k+1}u1^y have fiber Jacobian z^2, so they define a holomorphic family from W_k at t=0 to W2(y) at t=1. Thus the main construction is sound. The remaining weaknesses are expository: the text contains typos in the intermediate steps of Lemmas 1.9 and 1.10, and Section 2.2's moduli discussion is informal, especially around Corollary 2.22. These justify the existing CONDITIONAL verdict, but they are not load-bearing for the infinite-deformation claim.","tokens_in":20059,"tokens_out":58926,"duration_ms":562309,"concrete_test":"Run a symbolic computation (e.g., in Sage or Mathematica) of the Cech coboundary equation for W2(3): expand a general α ∈ O(U) and β ∈ O(V), impose σ_0 = [0, z^{-1}, 0]^T = α + T^{-1}β, and solve for the Laurent coefficients; confirm the system is inconsistent, and repeat for σ_p with p a polynomial of degree at most 1. This directly verifies the 'only for s ≥ y-1' claim in Theorem 1.12.","verdict_should_be":"UNCHANGED","load_bearing_attack":"After checking the central argument, no load-bearing concern is found. In Theorem 1.12, the assertion that monomials of the form z^{-1}u1^0u2^s appear only for s ≥ y-1 on the right-hand side of the coboundary equation is correct. The α contribution has nonnegative z-powers; the β1(2zu1+u2^y) contribution has u2-degree at least y, the β2 z^{-2} contribution has u2-degree at least y, and the β3 y z^{-1}u2^{y-1} contribution has u2-degree at least y-1, with equality only for the i=l, s=0 monomial. Thus no cancellation can produce z^{-1}u2^s for s < y-1. Lemma 1.9's reduction is also sound: after applying J to a general cocycle, only second-row terms with l=-1, i=0 survive with a negative ξ-power; all other terms become holomorphic on V. The step from Theorems 1.13 and 1.28 to Cor 1.29, though tersely stated, can be made explicit: pulling W2(y) back through the isomorphism of Theorem 1.28 and adding the deformation parameter t gives transition functions v1 = z^k u1 + t z^2 u2 + t^2 z u1^y, v2 = z^{-k+2}u2 + t z^{-k+1}u1^y, whose fiber Jacobian is exactly z^2, so the transition is biholomorphic for all t; at t=0 this is W_k and at t=1 it is W2(y). The remaining issues—typos in Lemmas 1.9 and 1.10 and the informal moduli discussion in Section 2.2—are expository and do not threaten the infinite-deformation claim.","agreement_with_reader":"disagree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper studies noncompact Calabi-Yau threefolds W_k = Tot(O_{P^1}(-k) ⊕ O_{P^1}(k-2)), k ≥ 1. The first half constructs explicit deformations W2(y) of W2 by transition functions v1 = z^2 u1 + z u2^y, v2 = u2, and computes H^1(W2(y), T W2(y)), proving that h^1 = y-1 for y ≥ 2 and hence that the W2(y) are pairwise non-isomorphic (Thm. 1.13). It then records a deformation family for W3, proves non-isomorphism of affine-bundle deformations W3(j), constructs holomorphic maps between W2 and W3, and states Cor. 1.29 that W_k has infinitely many distinct deformations for every k > 1. The second half studies rank-2 vector bundles on W_k, describes a 'generic part' of the moduli of extensions, and proves Theorem 2.21 that a deformation of W2 lowers the dimension of M2(W2) while keeping the moduli nontrivial, with Cor. 2.22 asserting that the W_k admit infinitely many deformations, some with nontrivial moduli.","tokens_in":20394,"tokens_out":25352,"duration_ms":222376,"significance":"If the main claims are correct, the paper gives explicit infinite families of non-isomorphic deformations of noncompact Calabi-Yau threefolds, contrasting with the surface case, and shows that deformations need not destroy moduli of vector bundles. The cohomology computation for the W2(y) is explicit and checkable, and the deformation families are given by concrete transition functions; these are genuine strengths. The moduli results, if made rigorous, would be an interesting addition. However, the paper currently mixes rigorous computations with informal statements about moduli spaces, and the step from Theorems 1.13 and 1.28 to Cor. 1.29 is only sketched.","major_comments":[{"comment":"The proof of Corollary 1.29 is not written out, and the cited ingredients do not directly imply the statement. Theorem 1.28 shows that for k > q > 0 the particular deformation W_k^q with v1 = z^k u1 + z^q u2 is isomorphic to W_q. For a fixed k there are only finitely many q < k, so combining Theorem 1.28 with Theorem 1.13 does not by itself yield infinitely many distinct deformations of that fixed W_k. What is needed is an explicit family with central fiber W_k and with fibers W2(y) for infinitely many y; the manuscript only sketches this in Example 1.26 and in the sentence preceding Proposition 1.27. Such a family can be written down (for k > 2, for example, v1 = z^k u1 + t z^2 u2 + t^2 z u1^y, v2 = z^{-k+2}u2 + t z^{-k+1}u1^y, which for t = 1 is isomorphic to W2(y) after the coordinate change of Theorem 1.28 with q = 2), and I recommend adding this computation. As it stands, Corollary 1.29 is a claim with a missing proof.","section":"Section 1.5, Cor. 1.29"},{"comment":"The paper defines M_j(W_k) as a quotient Ext^1_{W_k}(O(j), O(-j))/~ and then asserts that this quotient satisfies the definition of a coarse moduli space. The argument given in the paragraph after Problem 2.8 ('by upper semicontinuity every element near E_p can also be represented by an element of Ext^1') only shows the existence of local families of extension classes; it does not establish the existence of a scheme or analytic variety structure on the quotient, nor its corepresentability, nor that the dimension used later is well-defined. Since Theorem 2.21 compares dimensions of M2(W2) and M2(W2(τ)), the notion of dimension of these quotients needs a rigorous definition. If the intended meaning is the dimension of the locally closed subset of Ext^1 modulo the automorphism-group action, this should be stated explicitly and proved for the specific spaces used.","section":"Section 2.2, Eq. (12) and following paragraph"},{"comment":"The proof of Theorem 2.21 establishes that two specific bundles, corresponding to the classes z u1 and z u2, are non-isomorphic (using Lemma 2.20). This shows that the moduli set has at least two points, but it does not show that the moduli is positive-dimensional, which is what 'keeping the moduli nontrivial' and the introductory claim T3 require. To conclude a dimension drop from 3 to a positive dimension, one needs to show that a positive-dimensional family of non-isomorphic bundles exists in the deformed moduli, e.g. by proving that a generic linear combination of the remaining generators yields a family of non-isomorphic classes. Please either provide such an argument or weaken the claim to 'contains at least two distinct isomorphism classes'.","section":"Theorem 2.21, proof"}],"minor_comments":[{"comment":"The word 'explicitily' should be 'explicitly'.","section":"Abstract"},{"comment":"In the displayed expression for Jσ after the coordinate change, the factor should be (ξ^2 v1 − ξ v2^y)^i, not (ξ^2 v2 − ξ v2^y)^i.","section":"Lemma 1.9, proof"},{"comment":"The displayed equality in Case 3 should be 2z u2^s(z u1 + 1) (equivalently 2u2^s(z^2 u1 + z)), not 2u2^s(z^2 u1 + 1); the final conclusion is unaffected.","section":"Lemma 1.10, Case 3"},{"comment":"In the second entry of the coboundary matrix, the first summand should be 3β1 z u1 rather than 3β1 z^3 u1, and the sign in the third entry should be checked; the stated non-vanishing conclusion is not affected because the β1 terms contain a positive power of u1.","section":"Lemma 1.18, proof"},{"comment":"The notation M_j(W_k) is used for several different objects (the full quotient, the first-order subset, and the projectivized open part), with M(W_i; j) also appearing; please introduce distinct notation for these objects to make the dimension statements easier to follow.","section":"Section 2.2, Notation 2.5 and Lemma 2.6"}],"recommendation":"major_revision","confidential_remarks":"The core deformation computation in Section 1 appears checkable and is likely correct; the paper's main gap on the deformation side is the missing explicit family behind Cor. 1.29, which is fixable within the scope of the manuscript. The moduli section is more substantially under-developed: the existence and dimension of the quotients M_j(W_k) and M_j(W2(τ)) are not rigorously established, and Theorem 2.21 proves less than its statement. If the authors can add the missing family and either prove or clearly define the moduli-dimension claims, the paper would be a solid contribution to the deformation theory of noncompact Calabi-Yau threefolds."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear colleague,\n\nQuick take: the paper does what it says on the deformation side. The W2(y) family really gives infinitely many pairwise non-isomorphic noncompact Calabi-Yau threefolds, and Cor 1.29 follows from Theorems 1.13 and 1.28. I checked the key step in Theorem 1.12 because that is the load-bearing computation, and the stress-test note is right: monomials z^{-1}u2^s cannot appear on the coboundary side for s < y-1. The alpha term is holomorphic, the beta1 term needs nonnegative z-power, the beta2 term has u2-degree at least y, and the beta3 term has degree at least y-1. No hidden cancellation produces the missing monomials. So the main invariant h^1(W2(y), TW2(y)) = y-1 is credible, and the pairwise non-isomorphism is established.\n\nWhat is genuinely new: the h^1 computation as an isomorphism invariant, the affine-bundle classification for W3(j), and the transfer of infinite deformation families from W2 to all W_k with k > 1. The framework and the base family come from GKRS18, but the non-isomorphism results are new and are not circular. The citation pattern is fine; prior work is credited appropriately.\n\nThe soft spots are mostly in Section 2.2, and they are real but not fatal to the main theorem. The passage from quotients of Ext^1 to a coarse moduli space is informal. The text says a small disk around each extension class implies the naive quotient satisfies the definition of coarse moduli, but that is not a proof, and upper semicontinuity does not do the work by itself. Theorem 2.21 also assumes that the deformed Ext^1 is generated by the same four cocycles with only the relations exhibited; Example 2.18 sketches this but does not fully check whether additional generators vanish or extra relations appear. Lemma 2.20 is a solid concrete computation, but it only shows two classes are distinct, which supports nontriviality without giving a full moduli-space construction. Corollary 2.22 is stated without proof; it probably follows by combining Theorem 2.21 with Theorem 1.28, but it needs an explicit argument.\n\nThere are also minor typos in Lemmas 1.9 and 1.10 and a few rough edges in the exposition, but none of these threaten the central deformation result. The Laurent-series argument in Theorem 1.12 is terse, and the authors should expand it for the referee, but it is correct as written.\n\nWho is this for? People working on local Calabi-Yau threefolds, deformation theory of noncompact manifolds, or moduli of bundles on local spaces will get real use out of it. It deserves a serious referee. My recommendation: send it to peer review, with the expectation of revision. The deformation half should be accepted; the moduli half needs tightening before publication.","headline":"The infinite-deformation claim is solid and the key cohomology computation survives close inspection; the moduli half is real but under-proved.","tokens_in":20981,"tokens_out":2776,"would_cite":true,"duration_ms":31393,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["32G05","32G08","32Q25"],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper constructs explicit gluing families showing that the noncompact Calabi-Yau threefold W_k has infinitely many pairwise non-isomorphic deformations for every k>1, and that some of these deformations retain nontrivial moduli…","keywords":["Calabi-Yau threefolds","deformation theory","noncompact complex manifolds","moduli of vector bundles","tangent bundle cohomology","total spaces of vector bundles","holomorphic vector bundles","projective line"],"falsifier":"Exhibit explicit holomorphic functions $\\alpha$ on $U=\\mathbb{C}^3$ and $\\beta$ on $V=\\mathbb{C}^3$ solving $\\sigma_s = \\alpha + T^{-1}\\beta$ for some $0\\le s\\le y-2$, where $\\sigma_s=(0,z^{-1}u_2^s,0)^T$; even one such solution would reduce $h^1(W_2(y),T W_2(y))$ below $y-1$, invalidating the claimed infinitude of deformations.","tokens_in":2220,"feed_emoji":"♾️","tokens_out":3753,"duration_ms":101860,"temperature":0.7,"pith_summary":"This paper studies noncompact Calabi-Yau threefolds built as total spaces of rank-two bundles on the projective line, specifically $W_k=\\operatorname{Tot}(\\mathcal{O}_{\\mathbb{P}^1}(-k)\\oplus\\mathcal{O}_{\\mathbb{P}^1}(k-2))$. It aims to show that, unlike the surface case where commutative deformations destroy moduli of vector bundles, these threefolds admit infinitely many distinct deformation types, and some of those deformations still carry positive-dimensional moduli spaces of holomorphic vector bundles. The central construction is an explicit family $W_2(y)$ given by a gluing formula with an integer parameter $y$, and the paper proves $W_2(y_1)\\cong W_2(y_2)$ only when $y_1=y_2$, yielding infinitely many non-isomorphic deformations of $W_2$. A vector-bundle extension trick then transfers this infinitude to every $W_k$ with $k>1$. If correct, the paper establishes that noncompact Calabi-Yau threefolds can have very large deformation spaces, in contrast to the formally rigid case $W_1$.","feed_headline":"Calabi-Yau threefold Wk has infinitely many distinct deformations","feed_subtitle":"Explicit gluing formulas give pairwise non-isomorphic W2(y), and extension bundles spread them to every k>1.","key_machinery":"The load-bearing object is the family of threefolds $W_k$ with canonical gluing of two copies of $\\mathbb{C}^3$ by $(\\xi,v_1,v_2)=(z^{-1}, z^k u_1, z^{-k+2}u_2)$, together with deformations obtained by modifying the middle coordinate. For $W_2$, the family $W_2(y)$ is defined by the gluing $v_1 = z^2u_1 + z u_2^y$; the argument rests on a direct Čech computation of $H^1(W_2(y), T W_2(y))$, whose generators are cocycles $\\sigma_s = (0, z^{-1}u_2^s,0)^T$ and whose dimension $y-1$ is shown by analyzing which monomials can appear as coboundaries. For general $k$, the key mechanism is an extension family of rank-two bundles on $\\mathbb{P}^1$ with transition matrix $\\begin{pmatrix} z^k & z^q \\\\ 0 & z^{-k+2}\\end{pmatrix}$, whose total-space deformation is isomorphic to $W_q$; this transfers deformations from lower to higher $k$.","core_discovery":"The core discovery is that the threefold $W_k=\\operatorname{Tot}(\\mathcal{O}_{\\mathbb{P}^1}(-k)\\oplus\\mathcal{O}_{\\mathbb{P}^1}(k-2))$ has infinitely many pairwise non-isomorphic deformations whenever $k>1$. For $k=2$, the paper writes deformations $W_2(y)$ by the gluing $(\\xi,v_1,v_2)=(z^{-1}, z^2u_1 + z u_2^y, u_2)$ and computes $h^1(W_2(y), T W_2(y)) = y-1$ for $y\\ge2$; since this cohomology dimension is an isomorphism invariant, the $W_2(y)$ give infinitely many distinct complex structures. A family of rank-two bundles over $\\mathbb{P}^1$ interpolating between $W_k$ and $W_q$ then shows that the deformation of $W_k$ given by $(z^{-1}, z^k u_1 + z^q u_2, z^{-k+2}u_2)$ is isomorphic to $W_q$, so the infinite family for $W_2$ induces infinitely many deformations for every $k>1$. The paper also shows that some of these deformations preserve nontrivial moduli of vector bundles, with dimension decreasing but not collapsing to a point.","pith_inferences":["One consequence the paper leaves implicit is that the deformation space of a noncompact Calabi-Yau threefold can have at least countably many distinct components, in contrast to the finite-dimensional and usually connected deformation spaces of compact Calabi-Yau threefolds.","The extension-family argument suggests a hierarchy among the $W_k$: each $W_k$ can be deformed down to any $W_q$ with $0<q<k$, so deformation defines a partial order that could be visualised as a directed chain of threefolds.","A natural independent test of the non-isomorphism claim would be to compute higher cohomological or Hodge-theoretic invariants of $W_2(y)$; any such invariant depending on $y$ would corroborate Theorem 1.13 without relying solely on the tangent-cohomology calculation.","The paper's restriction to first-order extension classes when defining moduli spaces leaves open the behaviour of higher-order terms; understanding them could connect this construction to BPS-state counting on toric Calabi-Yau threefolds, a motivation the paper mentions but does not develop."],"forward_implications":["For $y\\ge2$, the threefolds $W_2(y)$ are pairwise non-isomorphic and non-affine, so a single noncompact Calabi-Yau threefold $W_2$ has infinitely many deformation classes.","For every $k>1$, $W_k$ admits infinitely many non-isomorphic deformations; in particular, the deformation with middle coordinate $z^k u_1 + z^q u_2$ is isomorphic to $W_q$ for $0<q<k$.","The threefold results invert the surface pattern: nontrivial deformations of $W_k$ need not be affine, and they are not obtained by deforming the compactification.","Some deformations of $W_2$ lower the dimension of the moduli space $M_2(W_2)$ from $3$ to at most $2$ without collapsing it, so positive-dimensional moduli of vector bundles can survive deformation.","Since $h^1(W_2(y),T W_2(y))=y-1$, the isomorphism classes in this family can be separated by an integer invariant, making the family a countably infinite set of distinct complex structures."],"supporting_citations":[{"why":"Supplies the definition of commutative deformation for noncompact manifolds and the infinite gluing family for W2 that the paper refines into W2(y).","marker":"[GKRS18]"},{"why":"Establishes the surface analogue — every nontrivial deformation of Z_k is affine and moduli of bundles collapse — which the threefold results are designed to contradict.","marker":"[BG19a]"},{"why":"Computes H^1(W1,T W1)=0, the base case showing W1 is formally rigid and motivating the restriction to k>1.","marker":"[Rub17]"},{"why":"Provides the filtrability theorems and the dimension of the generic moduli M_j(W_k) used to compare moduli before and after deformation.","marker":"[K¨10]"},{"why":"Restricts the normal bundle of the contracted line, justifying the choice of W_k among possible threefold models.","marker":"[J+92]"},{"why":"Gives the semiuniversal deformation space parametrising cocycles for W3, used in the affine-bundle analysis of Section 1.3.","marker":"[GKMR12]"},{"why":"Supplies the compact deformation-theory paradigm of computing H^1 with tangent coefficients and then integrating, which the noncompact construction follows.","marker":"[Kod06]"},{"why":"Provides the isomorphism between Ext^1 and H^1 used to describe extension classes defining the vector bundles on the deformations.","marker":"[Har13]"}],"fun_headline_variants":["Calabi-Yau W_k: infinitely many distinct deformations for every k>1","Explicit gluing gives infinite non-isomorphic deformations of W_k","From W_2 to all W_k: infinite Calabi-Yau deformations via bundles","W_k has infinite pairwise distinct deformations, proven explicitly","Infinite family of Calabi-Yau threefolds from each W_k"],"cache_read_input_tokens":22912,"weakest_assumption_plain":"The argument assumes that a Laurent-series analysis of the coboundary equation is exhaustive: no holomorphic functions on the two charts can combine to cancel into the monomials $z^{-1}u_2^s$ for $s<y-1$. If such cancellation existed, the dimension count $h^1(W_2(y),T W_2(y))=y-1$ and the pairwise non-isomorphism of the $W_2(y)$ would collapse.","fun_headline_variants_meta":{"raw":{"variants":["Calabi-Yau W_k: infinitely many distinct deformations for every k>1","Explicit gluing gives infinite non-isomorphic deformations of W_k","From W_2 to all W_k: infinite Calabi-Yau deformations via bundles","W_k has infinite pairwise distinct deformations, proven explicitly","Infinite family of Calabi-Yau threefolds from each W_k"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001768,"raw_usage":{"total_tokens":6937,"prompt_tokens":870,"completion_tokens":6067,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":486,"completion_tokens_details":{"reasoning_tokens":5966}},"tokens_in":486,"tokens_out":6067,"duration_ms":37159,"temperature":1.0,"reasoning_tokens":5966,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T05:09:22.394553+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Exhibit explicit holomorphic functions $\\alpha$ on $U=\\mathbb{C}^3$ and $\\beta$ on $V=\\mathbb{C}^3$ solving $\\sigma_s = \\alpha + T^{-1}\\beta$ for some $0\\le s\\le y-2$, where $\\sigma_s=(0,z^{-1}u_2^s,0)^T$; even one such solution would reduce $h^1(W_2(y),T W_2(y))$ below $y-1$, invalidating the claimed infinitude of deformations.","supporting_citations":[],"review_version":1}