{"id":"15252a75-1b69-4022-a665-cc43598993f1","arxiv_id":"2506.13546","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"On nilmanifolds with nilpotent or holomorphically parallelizable complex structures, p-Kähler structures are shown to imply (p+1)-Kähler (or balanced) structures, with new deformation obstructions and cohomological criteria.","lead":"Working in complex geometry, this paper proves two new cases of the Alessandrini-Bassanelli conjecture, which predicts that p-Kähler structures always force stronger (p+1)-Kähler structures. It also identifies a cohomological obstruction to deforming p-Kähler structures in families, with explicit examples.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Theorem 4.1 induction assumes a p-Kähler form on Γ\\G descends to the Lie subalgebra k; this is unproved and load-bearing.","rationale":"The reader's weakest_assumption focused on the characterization [6, Theorem 3.2] and on [43, Theorem 2.3]. I think the more delicate point is upstream: the descent of a p-Kähler structure from M to the Lie subalgebra k. That is exactly where non-invariance matters and where a global differential form on a compact quotient must be converted into an algebraic datum on a Lie algebra. If this conversion is invalid, no amount of correctness in [6] or [43] saves Theorem 4.1. It also shows that the rationality hypothesis is not the only bottleneck: even with Q[i] coefficients, the induction could fail at the descent step. The concern is external in the sense that the key fact is only cited to a proof in another paper; the present paper does not state an averaging or descent theorem. I therefore partially agree with the reader, but I identify a different specific step. I recommend keeping the verdict CONDITIONAL: the paper is plausible and the examples are interesting, but the main induction should not be accepted until the descent from M to k is either proved or verified against [18, Proposition 3.3] under the stated non-invariant hypothesis.","tokens_in":21475,"tokens_out":26067,"duration_ms":255446,"concrete_test":"Read the proof of [18, Proposition 3.3] and determine its hypotheses: does it start from an arbitrary (possibly non-left-invariant) p-Kähler form on Γ\\G, or from a left-invariant representative on the Lie algebra? Then run the following computational check: on the 6-dimensional holomorphically parallelizable nilmanifold with dφ^3=φ^{12}, dφ^4=φ^{13}, dφ^5=φ^{14}, dφ^6=φ^{15} (Q[i] coefficients), take a non-invariant 2-Kähler form Ω and attempt the contraction/descent used in [18, Proposition 3.3] to produce a 1-Kähler form on k; verify d-closedness and transversality. If the construction only works for invariant Ω, the induction step is not justified for the stated theorem.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The induction in Theorem 4.1 turns the assumed p-Kähler structure on M into a (p−1)-Kähler structure on the codimension-one complex Lie subalgebra k spanned by φ^1,...,φ^{n-1}, citing only the proof of Proposition 3.3 in [18]. This descent is what makes the induction go: without a (p−1)-Kähler structure on k, the induction hypothesis and [43, Theorem 2.3] are never triggered. The asserted step is not evident from the manuscript. M=Γ\\G is a compact quotient, while k is a Lie algebra, and the subgroup K integrating k is not naturally a submanifold of M because the Malcev lattice in K need not lie in Γ. Therefore Ω cannot literally be restricted or pushed forward to a (p−1)-form on K/Γ_K. Moreover, Ω is not assumed left-invariant. If [18, Proposition 3.3] is a statement about p-Kähler structures on Lie algebras (i.e., about invariant forms) rather than about arbitrary forms on the nilmanifold, an additional symmetrization or averaging result is needed, and the paper supplies none. If that unstated equivalence fails, Theorem 4.1 collapses for every n≥6, and with it the claimed resolution of the Alessandrini–Bassanelli conjecture in this class.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript studies p-Kähler, p-symplectic, and p-pluriclosed structures on compact complex manifolds. Its headline results are Theorem 4.1 (a p-Kähler structure on a holomorphically parallelizable nilmanifold with rational structure equations implies a (p+1)-Kähler structure) and Theorem 4.2 ((n−2)-Kähler nilmanifolds with nilpotent complex structure are balanced), which together would resolve the Alessandrini–Bassanelli conjecture in these classes. Section 5 gives a deformation obstruction (Theorem 5.1 and Corollary 5.2), and Section 6 gives cohomological nonvanishing criteria for p-Kähler, p-symplectic, and p-pluriclosed classes on nilmanifolds. Several worked examples are provided, including a new 3-Kähler structure on ηβ5 and families of 3-Kähler and 3-symplectic nilmanifolds.","tokens_in":21724,"tokens_out":8698,"duration_ms":80657,"significance":"If the main theorems are correct, they settle a long-standing conjecture in a substantial class and reduce p-Kähler geometry to balanced geometry on nilpotent nilmanifolds. The deformation formula and the cohomological nonvanishing theorem are useful and appear sound, and the explicit examples are a strength of the paper. However, the proof of the induction step in Theorem 4.1 contains a serious unsupported descent, the iterative construction has an internal inconsistency, and the proof of Theorem 4.2 contains a division-by-zero gap. Because these issues are load-bearing for the headline results, the significance cannot be fully assessed until they are repaired.","major_comments":[{"comment":"The induction step depends on the assertion that the complex Lie algebra k, spanned by φ^1,...,φ^{n-1}, φ̄^1,...,φ̄^{n-1}, admits a (p−1)-Kähler structure, with citation only to the proof of [18, Proposition 3.3]. This is not justified in the manuscript. The p-Kähler form on M=Γ\\G is not assumed left-invariant, and the subgroup K integrating k does not automatically give a compact submanifold K/Γ_K of M with Γ_K⊂Γ, so the given form cannot be restricted or descended without an argument. Since the induction hypothesis is applied to Mk=Γ_K\\K, this step is load-bearing: without an invariant reduction or averaging theorem, the induction does not go through. The author should either state and prove the descent, or prove a version of [18, Proposition 3.3] for the Lie algebra that applies to arbitrary (not necessarily invariant) forms.","section":"§4, Theorem 4.1 (descent step before Eq. (6))"},{"comment":"The construction of ζ_k is internally inconsistent. Equation (10) says 'ζ_{k−1}=0 if and only if ζ_{k−1}=a_k φ^1∧...∧φ^{k−1}', but if ζ_{k−1}=0 then ζ_k=−φ^k∧ζ_{k−1}=0, contradicting the assertion ζ_k≠0 in (11). The preceding formulas also do not rule out the possibility that the form being multiplied by φ^j is already zero at an earlier stage. Moreover, the claim that ζ is left-invariant is asserted without proof: [6, Theorem 3.2] yields an arbitrary smooth ζ, and no symmetrization step is supplied. This part of the proof must be rewritten before the induction can be evaluated.","section":"§4, Theorem 4.1, equations (7)–(11)"},{"comment":"The coframe change ψ^2=φ^2−(B^2_{11}/B^3_{11})φ^3 divides by B^3_{11}; the case B^3_{11}=0 is not treated. Since the theorem claims a conclusion for every nilpotent complex structure, this is a genuine gap. In addition, the step applying [43, Theorem 2.3] to conclude that at least one dψ^j vanishes for j≥3, and then re-ordering to obtain dψ^1=dψ^2=dψ^3=0, should be spelled out with the precise statement of the cited theorem; as written, the simultaneous normalization is not evident.","section":"§4, Theorem 4.2, after Eq. (13)"}],"minor_comments":[{"comment":"There are several typographical errors, including 'hyphotesis' in the introduction and 'paralellizable' in Section 2; a careful proofreading pass is recommended.","section":"Throughout"},{"comment":"The map denoted eιφ|ιφ is typeset inconsistently, sometimes as 'eιφ|iφ' and sometimes with subscripts missing; please standardize the notation for the deformation isomorphism.","section":"§3, notation"},{"comment":"Reference [15] appears in the bibliography but does not appear to be cited in the text; please check whether it should be cited or removed.","section":"References"},{"comment":"The identity eΩ_t=e^{iφ(t)|iφ(t)}(Ω*) is asserted without computation; since the example illustrates the obstruction criterion, a short verification or explicit reference would improve clarity.","section":"Example 5.3"}],"recommendation":"major_revision","confidential_remarks":null},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The short version: the paper proves two cases of the Alessandrini-Bassanelli conjecture on nilmanifolds and supplies a good set of new examples, but the main induction in Theorem 4.1 has a load-bearing gap that needs to be laid out explicitly, plus a couple of repairable typos.\n\nWhat is actually new: Theorem 4.1 gives the first proof, under Q[i] coefficients, of a statement previously asserted without proof in [7]; Theorem 4.2 (n-2-Kähler implies balanced for nilpotent complex structures) is new; Theorem 5.1 and Corollary 5.2 give a necessary cohomological condition for smooth curves of p-Kähler structures. The examples are concrete and useful: a new 3-Kähler structure on ηβ5, a family of 3-Kähler nilmanifolds of dimension 5, a family of 3-symplectic nilmanifolds, and a deformation where the obstruction in Corollary 5.2 is computable. Section 6 has simple but clean cohomological vanishing criteria for p-Kähler, p-symplectic, and p-pluriclosed structures.\n\nWhere the soft spots are. Theorem 4.1's induction hinges on the statement that the codimension-one subalgebra k admits a (p-1)-Kähler structure, cited to [18, Proof of Proposition 3.3]. This needs to be a real descent theorem: from a p-Kähler form on the compact nilmanifold Γ\\G to a left-invariant structure on the Lie algebra k. The manuscript gives no argument, and it is not obvious because Ω is not assumed invariant and K is not naturally a submanifold of Γ\\G. If [18] contains the needed averaging/descent result, the author should state it explicitly; otherwise the induction does not go through for n≥6. Also, the ζ iteration has a typo: once ζ1,...,ζ_{k-1} vanish, ζ is a multiple of φ1∧...∧φ^{k-1}, so ζ_{k-1}=0; defining ζ_k = -φ^k∧ζ_{k-1} then gives zero. The fix is to use -φ^k∧ζ directly. In Theorem 4.2, the coframe change divides by B^3_{11}; the vanishing case needs separate treatment. These are repairable, but they need to be fixed.\n\nThe paper leans heavily on external characterizations [6,18,22,43]; I did not verify them in the original sources, so I cannot guarantee the foundations. That is normal practice in this area, but the referee should check those dependencies.\n\nBottom line: this is a paper for people working on non-Kähler and p-Kähler geometry. The conjectures are central, the examples are valuable, and the gaps are fixable. Send it to peer review, with a clear request that the descent step and the zero-coefficient case be repaired and the typos cleaned up.","headline":"Worth reviewing: proves two cases of the Alessandrini-Bassanelli conjecture on nilmanifolds with good new examples, but Theorem 4.1 has a real gap in the induction that needs to be fixed.","tokens_in":22250,"tokens_out":11363,"would_cite":true,"duration_ms":101776,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["32J27","53C15","32G05"],"pacs":[],"model":"deepseek-v4-flash","headline":"The paper proves the Alessandrini–Bassanelli conjecture for holomorphically parallelizable nilmanifolds with Gaussian rational structure constants and for (n-2)-Kähler nilpotent nilmanifolds, collapsing intermediate p-Kähler geometry to…","keywords":["p-Kähler structure","p-symplectic structure","p-pluriclosed structure","nilmanifold","nilpotent complex structure","holomorphically parallelizable","balanced metric","deformation of complex structure"],"falsifier":"The decisive check is to search the families in Examples 4.6 and 4.7 for a 5-dimensional nilpotent nilmanifold that admits a transverse closed real $(3,3)$-form while its structure equations cannot be rearranged into the form $d\\psi^1=d\\psi^2=d\\psi^3=0$. Theorem 4.2 declares this impossible, so one such example would settle the claim against it.","tokens_in":21271,"feed_emoji":"📐","tokens_out":14109,"duration_ms":112517,"temperature":0.7,"pith_summary":"This paper attacks the Alessandrini–Bassanelli conjecture, which predicts that on a compact complex manifold a $p$-Kähler structure (a closed transverse $(p,p)$-form) implies a $(p+1)$-Kähler structure. It proves the conjecture for holomorphically parallelizable nilmanifolds whose structure constants lie in $\\mathbb{Q}[i]$, and it proves the top case for all nilpotent complex structures: a nilmanifold carrying an $(n-2)$-Kähler structure is balanced. If the arguments are right, the intermediate $p$-Kähler hierarchy in these classes collapses into balanced geometry. The paper also derives a necessary condition for a family of $p$-Kähler structures to move in a differentiable family of complex manifolds, and it shows the cohomology classes of $p$-Kähler, $p$-symplectic, and $p$-pluriclosed structures on invariant nilmanifolds are never zero.","feed_headline":"p-Kähler forms on nilmanifolds are balanced geometry in disguise","feed_subtitle":"A proof of the Alessandrini–Bassanelli conjecture for two nilmanifold classes, plus new deformation obstructions.","key_machinery":"The argument runs on left-invariant complex structure equations on nilpotent Lie algebras, $d\\varphi^j = \\sum_{i<k<j} A^j_{ik}\\varphi^{ik}$ (with possible mixed $B$ terms in the general nilpotent case), together with the external criterion that a compact complex manifold admits no $p$-Kähler structure exactly when there is a non-zero simple $(n-p,0)$-form that is $\\bar\\partial$-closed and $\\partial$-exact. The induction step passes to the intermediate Lie algebra $\\mathfrak{k}$ spanned by the first $n-1$ forms, which inherits a $(p-1)$-Kähler structure and, by Malcev's theorem under the rationality hypothesis, a lattice; the induction hypothesis then produces enough closed $(1,0)$-forms to raise the order of the form. In the $(n-2)$-Kähler case the machinery fixes a coframe with $d\\psi^1=d\\psi^2=d\\psi^3=0$ and builds a transverse $(n-1)$-form by wedging the given $\\Omega$ with $\\psi^{1\\bar1}$ and $\\psi^{2\\bar2}$. For deformation questions the paper uses the simultaneous contraction operator $e^{\\iota_\\varphi|\\iota_{\\bar\\varphi}}$ to pull forms on nearby fibers back to the central fiber and differentiates the condition $\\partial_t\\Omega_t=0$.","core_discovery":"The paper's central claim is Theorem 4.2: a compact nilmanifold of complex dimension $n$ with nilpotent complex structure admits a balanced metric as soon as it admits an $(n-2)$-Kähler structure. Theorem 4.1 is the fuller version for holomorphically parallelizable nilmanifolds: with structure equations $d\\varphi^j = \\sum_{i<k<j} A^j_{ik}\\varphi^{ik}$ and $A^j_{ik}\\in\\mathbb{Q}[i]$, any $p$-Kähler structure forces a $(p+1)$-Kähler structure. The proof is an induction on dimension: one passes to an intermediate complex Lie algebra $\\mathfrak{k}$ that inherits a $(p-1)$-Kähler structure, uses Malcev's theorem to give it a lattice, applies the induction hypothesis, and then invokes an external criterion saying that failure to admit a $(p+1)$-Kähler structure is equivalent to the existence of a non-zero simple $\\bar\\partial$-closed and $\\partial$-exact $(n-p-1,0)$-form. The final contradiction upgrades the form along the missing directions. For the $(n-2)$ case, the construction of the balanced form is explicit: $\\Omega_1 = \\sigma_1(\\Omega\\wedge\\psi^{1\\bar1}+\\Omega\\wedge\\psi^{2\\bar2})$ is shown transverse by a pointwise check on simple $(1,0)$-forms.","pith_inferences":["If the same induction could run without the rationality hypothesis, the collapse to balanced would follow for all holomorphically parallelizable nilmanifolds; the $\\mathbb{Q}[i]$ condition enters only through the lattice existence step, so removing it is the natural next target.","The deformation obstruction yields a practical stability test: compute whether $\\partial\\circ\\iota_{\\varphi'(0)}(\\Omega)$ is $\\bar\\partial$-exact; directions where it is not cannot carry any smooth curve of $p$-Kähler structures, which places the known instability examples in a systematic framework.","The non-vanishing theorem for nilmanifolds contrasts with known balanced metrics on other compact complex manifolds whose top power has zero cohomology class, so nilmanifolds appear to be the sharp regime for this cohomological obstruction."],"forward_implications":["On holomorphically parallelizable nilmanifolds with structure constants in $\\mathbb{Q}[i]$, the $p$-Kähler hierarchy collapses: a $p$-Kähler structure forces all higher $q$-Kähler structures up to $q=n-1$, i.e. a balanced metric.","On nilpotent nilmanifolds, an $(n-2)$-Kähler structure is automatically a balanced metric, so the conjectured inclusion of $(n-2)$-Kähler into balanced holds in this class.","Any smooth curve of $p$-Kähler structures on a fixed central fiber must satisfy $\\partial\\circ\\iota_{\\varphi'(0)}(\\Omega)=-\\partial\\Omega'(0)$; in particular the class $[\\partial\\circ\\iota_{\\varphi'(0)}(\\Omega)]$ vanishes in $H^{p,p+1}_{\\bar\\partial}(M)$, giving a computable deformation obstruction.","On nilmanifolds with invariant complex structures, no $p$-Kähler, $p$-symplectic, or $p$-pluriclosed structure is cohomologically trivial in its natural cohomology.","The paper's examples give a new $3$-Kähler structure on the holomorphically parallelizable nilmanifold $\\eta^5_\\beta$, a family of $3$-Kähler nilmanifolds, a family of $3$-symplectic nilmanifolds, and a family of $3$-Kähler structures moving along a deformation."],"supporting_citations":[{"why":"Supplies the criterion that absence of a $p$-Kähler structure is equivalent to a non-zero simple $\\bar\\partial$-closed $\\partial$-exact $(n-p,0)$-form, the engine of the induction.","marker":"[6]"},{"why":"Theorem 2.3 there guarantees an $(n-2)$-Kähler nilpotent nilmanifold has a coframe with $d\\psi^1=d\\psi^2=d\\psi^3=0$, the starting point of Theorem 4.2.","marker":"[43]"},{"why":"Provides the base case in complex dimension 5 and the step that produces a $(p-1)$-Kähler structure on the intermediate Lie algebra.","marker":"[18]"},{"why":"Malcev's theorem supplies a lattice for the intermediate nilpotent Lie algebra with rational structure constants, making the induction manifold compact.","marker":"[29]"},{"why":"Proposition 3.4 is used to kill the leading structure coefficients in the $(n-2)$-Kähler proof.","marker":"[22]"},{"why":"Gives the positivity criteria used to prove the newly constructed transverse forms are indeed transverse.","marker":"[17]"},{"why":"Shows every left-invariant metric on a holomorphically parallelizable nilmanifold is balanced, covering the low-dimensional cases.","marker":"[1]"},{"why":"Supplies the canonical left-invariant basis with $d\\varphi^j$ in the ideal generated by earlier forms, used to prove non-vanishing of cohomology classes on nilmanifolds.","marker":"[40]"},{"why":"Identifies holomorphically parallelizable compact complex manifolds with quotients of complex Lie groups by lattices, grounding the nilmanifold setup.","marker":"[47]"}],"fun_headline_variants":["p-Kähler nilmanifolds are balanced: conjecture proved","Alessandrini–Bassanelli conjecture proven for two nilmanifold classes","Induction shows p-Kähler forces (p+1)-Kähler on nilmanifolds","Deformation obstructions for curves of p-Kähler structures","Cohomology of p-Kähler, p-symplectic, p-pluriclosed forms"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is twofold: the previously established criterion that a manifold has no $p$-Kähler structure exactly when a certain exact holomorphic form exists, and the rationality of the structure constants ($A^j_{ik}\\in\\mathbb{Q}[i]$) that makes the intermediate Lie algebra admit a lattice; if either premise is wrong, the induction in Theorem 4.1 collapses.","fun_headline_variants_meta":{"raw":{"variants":["p-Kähler nilmanifolds are balanced: conjecture proved","Alessandrini–Bassanelli conjecture proven for two nilmanifold classes","Induction shows p-Kähler forces (p+1)-Kähler on nilmanifolds","Deformation obstructions for curves of p-Kähler structures","Cohomology of p-Kähler, p-symplectic, p-pluriclosed forms"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000253,"raw_usage":{"total_tokens":1613,"prompt_tokens":1045,"completion_tokens":568,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":661,"completion_tokens_details":{"reasoning_tokens":456}},"tokens_in":661,"tokens_out":568,"duration_ms":4954,"temperature":1.0,"reasoning_tokens":456,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T20:00:31.442030+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"The decisive check is to search the families in Examples 4.6 and 4.7 for a 5-dimensional nilpotent nilmanifold that admits a transverse closed real $(3,3)$-form while its structure equations cannot be rearranged into the form $d\\psi^1=d\\psi^2=d\\psi^3=0$. Theorem 4.2 declares this impossible, so one such example would settle the claim against it.","supporting_citations":[{"cited_title":"Alessandrini, G","cited_arxiv_id":null,"evidence_quote":"Supplies the criterion that absence of a $p$-Kähler structure is equivalent to a non-zero simple $\\bar\\partial$-closed $\\partial$-exact $(n-p,0)$-form, the engine of the induction."},{"cited_title":"Sferruzza, N","cited_arxiv_id":null,"evidence_quote":"Theorem 2.3 there guarantees an $(n-2)$-Kähler nilpotent nilmanifold has a coframe with $d\\psi^1=d\\psi^2=d\\psi^3=0$, the starting point of Theorem 4.2."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the base case in complex dimension 5 and the step that produces a $(p-1)$-Kähler structure on the intermediate Lie algebra."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Malcev's theorem supplies a lattice for the intermediate nilpotent Lie algebra with rational structure constants, making the induction manifold compact."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Proposition 3.4 is used to kill the leading structure coefficients in the $(n-2)$-Kähler proof."},{"cited_title":"Abbena, A","cited_arxiv_id":null,"evidence_quote":"Shows every left-invariant metric on a holomorphically parallelizable nilmanifold is balanced, covering the low-dimensional cases."},{"cited_title":"Salamon, Complex structures on nilpotent Lie algebras,J","cited_arxiv_id":null,"evidence_quote":"Supplies the canonical left-invariant basis with $d\\varphi^j$ in the ideal generated by earlier forms, used to prove non-vanishing of cohomology classes on nilmanifolds."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Identifies holomorphically parallelizable compact complex manifolds with quotients of complex Lie groups by lattices, grounding the nilmanifold setup."}],"review_version":2}