{"id":"728b283d-25dd-424e-bec4-560b248e0011","arxiv_id":"1908.09045","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":3.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"A survey proposing a C-infinity locally trivial definition of noncompact complex deformations, with examples and Hodge/KKP diamonds from the authors' earlier work.","lead":"This paper proposes a definition for how noncompact complex shapes can be deformed smoothly, and then surveys examples from Calabi-Yau threefolds, cotangent bundles, and adjoint orbits, including diagrams of their Hodge-theoretic patterns. It is mostly a review of the authors' earlier results, so the new content is a single definition plus a collection of examples.","discovery_kind":"review","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The displayed KKP diamond for LG(3) in Section 5.3.2 is quoted from [BGRSM] without specifying a tame compactification or proving independence from that choice, despite Remark 5.2 documenting compactification dependence for adjoint orbits.","rationale":"I read the paper as a survey whose advertised content is a new deformation notion and a collection of Hodge and KKP diamonds for noncompact Calabi–Yau-type examples. The definition itself is plausible and the examples from prior work are real evidence. The weakest point is precisely the compactification dependence of the KKP invariants, and the paper itself flags this in Remark 5.2. Since the KKP diamond for LG(3) is quoted from a self-cited preprint without specifying the tame compactification or proving independence, the displayed diamond cannot currently be accepted as an invariant of the LG model. This matches the reader's conditional verdict: the central claims are plausible but require an additional check before the displayed diamonds can be taken at face value. I do not see a stronger objection that would justify rejection, and I do not see a reason to upgrade to acceptance without addressing the compactification issue.","tokens_in":12579,"tokens_out":4287,"duration_ms":44066,"concrete_test":"Construct two different tame compactifications of the affine orbit O3 = Ad(SL(3,C))·Diag(2,-1,-1), for example by homogenizing its defining ideal with respect to two different weight gradings as in [BCG], and compute the three KKP invariants f^{p,q}, h^{p,q}, and i^{p,q} for each compactification. Compare the resulting diamonds with each other and with the diamond displayed in Section 5.3.2; if any entry differs, the displayed KKP diamond is not a well-defined invariant of LG(3). A minimal first step is to write down the tame compactification used in [BGRSM, Sec. 7] explicitly and independently recompute the middle entry 3 from Definition 5.3, 5.4, and 5.5.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central advertised invariant in this paper is the KKP diamond, and the load-bearing step is Section 5.3.2: the paper states that 'LG(3) admits a tame compactification' and then displays a KKP diamond calculated in [BGRSM, Sec. 7]. Definition 5.1 does not assert uniqueness of a tame compactification, and Remark 5.2 explicitly notes that for adjoint orbits different homogenizations of the defining ideal can change Hodge-theoretic invariants drastically, giving h^{1,4}=h^{4,1}=16 in one compactification and h^{1,4}=h^{4,1}=1 in another. The paper neither fixes a particular tame compactification for LG(3) nor proves that the f^{p,q}, h^{p,q}, and i^{p,q} numbers are independent of such a choice. If the KKP diamond varies with the compactification, then the diamond displayed in Section 5.3.2 is not an invariant of the Landau–Ginzburg model (O3, fH), and the paper's claim that KKP diamonds are invariants of these LG models is unsupported. This is not a fatal objection to Definition 2.3 or to the deformation examples, but it is a genuine condition on the paper's main advertised computations.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proposes a new definition of deformation of complex structure for noncompact complex manifolds: a holomorphic surjective submersion to a disc whose central fiber is the given manifold and which is locally trivial in the C-infinity category (Definition 2.3). It contrasts this with the classical Kodaira definition, gives an example showing that a naive noncompact extension is too permissive, and then presents examples including the local surfaces Z_k = Tot(O_{P^1}(-k)), the threefolds W_k = Tot(O(-k) ⊕ O(k-2)), and semisimple adjoint orbits viewed as deformations of cotangent bundles of flag manifolds. The paper also recalls the Katzarkov-Kontsevich-Pantev (KKP) invariants for Landau-Ginzburg models, states the KKP conjecture, and displays Hodge diamonds and KKP diamonds for minimal adjoint orbits of sl(2,C), sl(3,C), and sl(n,C), quoting the KKP computations from the authors' earlier work [BGRSM].","tokens_in":12756,"tokens_out":6452,"duration_ms":68792,"significance":"The proposed Definition 2.3 is coherent, and Example 2.1 does illustrate a genuine pathology of the naive noncompact version of Kodaira's definition; the examples in Sections 2 and 4 do show that the definition is satisfied by interesting families, including adjoint orbits as deformations of cotangent bundles. If the KKP diamonds were shown to be independent of the choice of tame compactification, the paper would provide a useful survey of Hodge-theoretic invariants for noncompact Calabi-Yau and Landau-Ginzburg models. However, the paper's main advertised computations, the KKP diamonds, are quoted from a preprint that is not independently verified here, and the paper's own Remark 5.2 acknowledges that Hodge-theoretic invariants of adjoint orbits can change drastically with the choice of compactification. The central invariant claim is therefore not yet supported as stated. The paper is also not self-contained as a deformation theory: no structural theorem (existence, versality, or completeness) is proved for Definition 2.3, so its claimed advantage over the naive definition rests on the examples alone.","major_comments":[{"comment":"The displayed KKP diamond for LG(3) is not established as an invariant of the Landau-Ginzburg model (O_3, f_H). The paper asserts that 'LG(3) admits a tame compactification' and then quotes the diamond from [BGRSM, Sec. 7], but it does not specify which tame compactification is used, and Definition 5.1 does not imply uniqueness of a tame compactification. Remark 5.2 explicitly reports that for adjoint orbits, different homogenizations of the defining ideal can change Hodge-theoretic invariants drastically, giving h^{1,4}=h^{4,1}=16 in one compactification and h^{1,4}=h^{4,1}=1 in another. Consequently, the paper does not show that the f^{p,q}, h^{p,q}, and i^{p,q} numbers in the displayed diamond are independent of the compactification. Since the KKP diamond is a central advertised output of the paper, the authors must either specify the tame compactification used in [BGRSM], prove invariance of the diamond under all tame compactifications, or explicitly state that the diamond is a compactification-dependent computation rather than an invariant of the pair (O_3, f_H).","section":"§5.3.2, Remark 5.2"},{"comment":"The paper claims that Definition 2.3 is an adaptation of Kodaira's theory 'better suited' to noncompact manifolds, but it proves no structural properties of this new notion. In particular, Remark 2.5 proposes to choose the dimension of the parameter space D to be h^1(X, T X) 'whenever possible', yet no theorem is proved or cited showing that this dimension is achievable or that the resulting family is complete or versal in the sense of Definition 2.3. The only demonstrated advantage over the naive definition is that Example 2.1 is excluded. The authors should state precisely which properties of Kodaira's theory are preserved by Definition 2.3 (for example, existence of semiuniversal families, the Kodaira-Spencer map, or unobstructedness) and either prove them or clearly mark them as open. Without such a statement, the claim that this is a deformation theory rather than a convenient class of examples is supported only by the examples, not by a general result.","section":"§2, Definition 2.3 and Remark 2.5"},{"comment":"The displayed Hodge diamonds for the noncompact manifolds contain entries labeled '∞', but the meaning of these entries is never defined. It is unclear whether they denote ordinary cohomology, compactly supported cohomology, Borel-Moore homology, or some other invariant. Since the paper uses these diamonds to compare O_2 with T^*P^1 and O_3 with T^*P^2, the lack of a convention makes it impossible for the reader to verify the comparisons. A sentence defining the cohomology theory used for the '∞' entries is needed.","section":"§5.3.1, §5.3.2"}],"minor_comments":[{"comment":"The running title in the header appears as 'famille s' with an unusual space; the abstract also contains several typographical inconsistencies. The paper would benefit from a careful proofreading pass.","section":"Title and abstract"},{"comment":"The parameter space D is described as 'any smooth manifold (even P1 or a disc)', while Definition 2.3 requires a complex disc; the example is meant to illustrate the naive definition, but the mismatch should be clarified explicitly.","section":"Example 2.1"},{"comment":"The same notation Z_k is used for the original surface and for a nontrivial deformation of it (e.g., 'Z_k contains no compact complex analytic curves'). This is confusing; a different symbol or a clear parenthetical clarification would help.","section":"§2.1"},{"comment":"The phrase 'these are then required to be integral' should presumably be 'integrable'; the intended meaning is clear but the typo should be corrected.","section":"§3, Remark 3.2"}],"recommendation":"major_revision","confidential_remarks":"The paper is largely a survey of the authors' own prior work, and the main new computations (the KKP diamonds) are quoted from [BGRSM], which is coauthored by two of the authors and is cited as an arXiv preprint. The missing specification of the tame compactification in §5.3.2 is therefore more consequential than it would be if the computation were independently reproduced here. I would encourage the authors either to include the relevant computation or to state explicitly that the displayed diamond corresponds to a particular compactification chosen in [BGRSM] and that the question of invariance is open. The paper also fits the journal's scope, but its contribution would be clearer if the deformation-theoretic claims were separated from the survey material."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear colleague,\n\nQuick take: this is a survey with a small conceptual contribution. Definition 2.3 — a holomorphic surjective submersion to a disc with central fiber X and C-infinity local triviality — is a reasonable adaptation of Kodaira's definition, and Example 2.1 does real work: it shows the naive noncompact version admits pathological families even when H^1(X,TX)=0. That motivates the definition. The rest of the paper collects the authors' previous results on Z_k and W_k surfaces/threefolds and on adjoint orbits, presented as Hodge and KKP diamonds. If you want an entry point into this cluster of work, this is a readable one.\n\nWhat's genuinely new is just the definition and the packaging. The examples and diamond computations come from [BG1], [R], [GS], [GGSM1], [BGRSM]. That's fine for a survey, but the abstract's \"we introduce a new notion\" overstates it: the definition is a modest reformulation of the usual locally trivial family condition, not a new deformation-theoretic framework. The authors don't prove that their condition is the right one beyond the examples.\n\nThe soft spot the stress-test flags is real and I agree it's the main issue. Section 5.3.2 states that LG(3) admits a tame compactification and displays a KKP diamond quoted from [BGRSM] without fixing which tame compactification. Definition 5.1 doesn't assert uniqueness, and Remark 5.2 explicitly documents cases where two homogenizations of the same adjoint orbit give different Hodge numbers (h^{1,4}=16 vs 1). So the displayed diamond is not shown to be an invariant of the Landau–Ginzburg model. The authors could patch this by specifying a particular compactification or by proving independence for these examples; as written, the central advertised computation is conditional. That doesn't sink Definition 2.3 or the deformation examples, but it should be fixed before the KKP diamonds are used as evidence.\n\nMinor point: the paper leans heavily on self-citations. That's natural for a survey but worth keeping in mind when evaluating the \"new\" content.\n\nBottom line: I'd send this to a referee. It's a clear survey with a coherent proposed definition and one honest caveat that needs addressing. A good referee could push for the compactification specification and for toning down the novelty claim. If you work on noncompact deformation theory, the definition is worth engaging with; otherwise it's a reasonable overview, not a must-read.","headline":"A clear survey with a reasonable working definition for deformations of noncompact manifolds, but the advertised KKP diamond isn't yet an invariant because the tame compactification is unspecified.","tokens_in":13351,"tokens_out":2115,"would_cite":false,"duration_ms":19884,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["32G05"],"pacs":[],"model":"deepseek-v4-flash","headline":"For noncompact manifolds, a deformation family should be a holomorphic surjective submersion to a disc that is locally trivial in the smooth category; the paper's examples show this definition captures genuinely new families.","keywords":["deformations of complex structures","noncompact Calabi–Yau manifolds","adjoint orbits","Landau–Ginzburg models","KKP conjecture","Hodge diamonds","cotangent bundles of flag manifolds","hyperkähler families"],"falsifier":"For the LG model on the four-dimensional adjoint orbit of $\\mathrm{Diag}(2,-1,-1)$ in $\\mathfrak{sl}(3,\\mathbb{C})$, compute $f^{p,q}$, $h^{p,q}$, and $i^{p,q}$ using two different ways of completing the model to a compact projective variety that satisfy the paper's Definition 5.1. If any entry of the resulting KKP diamond changes between the two completions, or if $f^{p,q}=h^{p,q}=i^{p,q}$ fails, then the diamond quoted for LG(3) is not an invariant of the model and the Hodge-theoretic claim would need revision.","tokens_in":12309,"feed_emoji":"💎","tokens_out":17514,"duration_ms":149013,"temperature":0.7,"pith_summary":"This paper argues that the classical deformation theory of compact complex manifolds is the wrong tool for noncompact ones: vanishing of $H^1(X,T_X)$ can coexist with nontrivial families, and the naive noncompact version lets the smooth type change. It proposes Definition 2.3, in which a deformation family of $X$ is a holomorphic surjective submersion $\\pi:\\widetilde X\\to \\mathbb{D}$ over a disc with $\\pi^{-1}(0)=X$ and with the family locally trivial in the $C^\\infty$ category. The supporting examples are concrete: the surfaces $Z_k=\\mathrm{Tot}(\\mathcal{O}_{\\mathbb{P}^1}(-k))$ have semiuniversal deformation families whose nontrivial members are affine and contain no compact curves, and the threefold $W_2$ has an infinite-dimensional family with infinitely many isomorphism types. The paper then computes Hodge diamonds for these spaces and for $T^*\\mathbb{P}^1$, $T^*\\mathbb{P}^2$, and adjoint orbits, and displays KKP diamonds for the associated Landau–Ginzburg models, where the three KKP invariants coincide. A reader should care because a working deformation theory for noncompact Calabi–Yau manifolds would control how complex structure, moduli of bundles, and Hodge-theoretic invariants change in families.","feed_headline":"Deform noncompact manifolds without changing their smooth type","feed_subtitle":"It keeps the smooth type fixed while changing complex structure and supports concrete Hodge and KKP diamond computations.","key_machinery":"The load-bearing object is the paper's proposed Definition 2.3, which replaces compactness in the classical analytic-family definition by local triviality in the $C^\\infty$ category: the smooth type of the manifold is fixed while the complex structure is allowed to vary, and pathological families such as Example 2.1 are excluded. For the adjoint-orbit examples, the engine is the paper's Theorem 3.4, which realizes $G^c/H^c$ as a family of complete hyperkähler metrics parametrized by triples $(\\tau_1,\\tau_2,\\tau_3)$ in a Cartan subalgebra with common stabilizer; the corollary used here is that a complex adjoint orbit is a deformation of the cotangent bundle of a flag manifold. For the diamond calculations, the machinery is the triple of KKP invariants—$f^{p,q}$ from logarithmic forms adapted to the superpotential, $h^{p,q}$ from relative cohomology with its weight filtration, and $i^{p,q}$ from vanishing cycles—together with the KKP conjecture that these three numbers coincide; the paper displays cases, e.g. the minimal adjoint orbits of $\\mathfrak{sl}(n,\\mathbb{C})$, where they do.","core_discovery":"The paper's central claim is that Definition 2.3 is the right adaptation of the classical deformation theory of compact complex manifolds to noncompact manifolds. A deformation family of $X$ is a holomorphic surjective submersion $\\pi:\\widetilde X\\to \\mathbb{D}$ with $\\pi^{-1}(0)=X$ and with $\\widetilde X$ locally trivial in the $C^\\infty$ category; the fibres $X_t=\\pi^{-1}(t)$ are the deformations. In this sense, the paper shows that each $Z_k$ admits a $(k-1)$-dimensional semiuniversal family whose nontrivial deformations are affine and contain no compact complex analytic curves, that $W_2$ has an infinite-dimensional family of deformations with both affine and non-affine members and infinitely many isomorphism types, and that a complex adjoint orbit $\\mathrm{Ad}(G)H_0$ is a deformation of the cotangent bundle of a flag manifold. On the Landau–Ginzburg side, using the potential $f_H(x)=\\langle H,x\\rangle$ on such an orbit, the paper records the Hodge diamonds of the orbits and cotangent bundles and the KKP diamonds of the LG models; for minimal adjoint orbits of $\\mathfrak{sl}(n,\\mathbb{C})$ the three KKP invariants $f$, $h$, $i$ coincide, so a KKP diamond is well defined there.","pith_inferences":["A natural next step is to compute the KKP diamonds on nontrivial deformations of $W_2$ or on deformations of the adjoint orbits; the paper leaves open how these diamonds vary in a family, and that computation would test whether KKP diamonds are deformation-invariant.","The compactification caveat suggests checking the quoted LG(3) diamond against two different tame compactifications; if it changes, the diamond belongs to the compactification, not to the Landau–Ginzburg model alone.","The same definition could be applied to other noncompact Calabi–Yau threefolds, such as total spaces of rank-two bundles over higher-genus curves, to see whether the affine/non-affine and formal-rigidity phenomena seen for $W_k$ persist."],"forward_implications":["Nontrivial deformations of the surfaces $Z_k$ are affine and contain no compact complex curves, and all holomorphic vector bundles on them split; moduli of instantons on the undeformed surface disappear after deformation.","The threefold $W_2$ has an infinite-dimensional, integrable deformation family with infinitely many pairwise non-isomorphic members, so noncompact Calabi–Yau threefolds can have very large deformation spaces even when related manifolds are formally rigid.","Every complex semisimple adjoint orbit is a deformation, in the new sense, of the cotangent bundle of a flag manifold, giving deformation families in every complex dimension.","For minimal adjoint orbits of $\\mathfrak{sl}(n,\\mathbb{C})$, the KKP equality $f^{p,q}=h^{p,q}=i^{p,q}$ holds, so the KKP conjecture is true in this family even though it is false in full generality.","Hodge diamonds can change between an orbit and the corresponding cotangent bundle (for example $O_3$ versus $T^*\\mathbb{P}^2$), so the new deformation notion preserves smooth type but not Hodge-theoretic invariants."],"supporting_citations":[{"why":"Supplies the classical definition of analytic family and the compact deformation theory that Definition 2.3 adapts.","marker":"[Ko]"},{"why":"Provides the semiuniversal deformation families for the surfaces $Z_k$ and the vector-bundle splitting results on their deformations.","marker":"[BG1]"},{"why":"Establishes formal rigidity of $W_1$ and the infinite-dimensional deformability of $W_2$ by constructing integrable families.","marker":"[R]"},{"why":"Shows there are infinitely many isomorphism types among the deformations of $W_2$ and that nontrivial deformations reduce moduli dimension.","marker":"[GS]"},{"why":"Supplies the hyperkähler families on $G^c/H^c$ that make adjoint orbits deformations of cotangent bundles of flag manifolds.","marker":"[Kv]"},{"why":"Provides the symplectic Lefschetz fibration structure on adjoint orbits via the potential $f_H$, used later for LG models.","marker":"[GGSM1]"},{"why":"Identifies the characteristic adjoint orbit as a $C^\\infty$ vector bundle isomorphic to $T^*F_\\Theta$.","marker":"[GGSM2]"},{"why":"Defines the three invariants $f^{p,q}$, $h^{p,q}$, $i^{p,q}$ and states the KKP conjecture.","marker":"[KKP]"},{"why":"Proves the KKP conjecture for minimal adjoint orbits and supplies the KKP diamonds displayed in the paper.","marker":"[BGRSM]"},{"why":"Gives an example where $i^{p,q}\\neq h^{p,q}$, showing the KKP conjecture fails in general and motivating the special families.","marker":"[LP]"}],"fun_headline_variants":["Smooth-type-preserving deformations of noncompact Calabi-Yau manifolds","Kodaira-style deformations that keep noncompact manifolds smooth","Affine families of noncompact CY deformations, Hodge diamonds computed","Noncompact deformation theory: new families, Hodge and KKP diamonds","Adjoint orbits as deformations of cotangent bundles, Hodge diamonds"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The paper assumes the KKP diamonds it displays are properties of the Landau–Ginzburg model alone, unchanged by the choice of how the model is completed to a compact variety at infinity, even though the paper's own caveat reports that Hodge numbers of the same orbit can change drastically with that choice.","fun_headline_variants_meta":{"raw":{"variants":["Smooth-type-preserving deformations of noncompact Calabi-Yau manifolds","Kodaira-style deformations that keep noncompact manifolds smooth","Affine families of noncompact CY deformations, Hodge diamonds computed","Noncompact deformation theory: new families, Hodge and KKP diamonds","Adjoint orbits as deformations of cotangent bundles, Hodge diamonds"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000695,"raw_usage":{"total_tokens":3128,"prompt_tokens":912,"completion_tokens":2216,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":528,"completion_tokens_details":{"reasoning_tokens":2128}},"tokens_in":528,"tokens_out":2216,"duration_ms":15027,"temperature":1.0,"reasoning_tokens":2128,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T11:23:34.991639+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"For the LG model on the four-dimensional adjoint orbit of $\\mathrm{Diag}(2,-1,-1)$ in $\\mathfrak{sl}(3,\\mathbb{C})$, compute $f^{p,q}$, $h^{p,q}$, and $i^{p,q}$ using two different ways of completing the model to a compact projective variety that satisfy the paper's Definition 5.1. If any entry of the resulting KKP diamond changes between the two completions, or if $f^{p,q}=h^{p,q}=i^{p,q}$ fails, then the diamond quoted for LG(3) is not an invariant of the model and the Hodge-theoretic claim would need revision.","supporting_citations":[],"review_version":1}