{"id":"289adc4f-de48-4f00-9b99-2c29fbf60241","arxiv_id":"2505.24702","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"On compact Hermitian manifolds whose metric satisfies ∂∂ω^k = 0 for all k, the relative non-pluripolar product of closed positive currents is well-defined and its mass is monotone.","lead":"This paper proves that a way of multiplying geometric currents, the relative non-pluripolar product, is always well-defined on a class of non-Kähler manifolds, extending an earlier result for Kähler manifolds. The result matters for studying complex Monge-Ampère equations and intersection theory on Hermitian manifolds.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The metric condition in Theorems 1.1 and 3.2 is type-inconsistent as stated; the integration-by-parts mass bound depends on the corrected condition ∂∂ω=0 and ∂ω∧∂̄ω=0.","rationale":"The reader's weakest assumption correctly identifies the metric condition as the crux. In good faith, the construction follows Vu's template, and the global mass bound is the essential new step. My check of the calculus shows that under the intended condition ∂∂ω=0, ∂ω∧∂̄ω=0, the identity ∫ ddc u ∧ S ∧ ω^q = ∫ u ∧ ddc(S∧ω^q) is valid because dS=0 and ddc(ω^q)=0 kill boundary terms, even when S contains other ddc factors. Thus the central claim appears mathematically sound. The remaining obstacle is that the manuscript states the condition incorrectly, making the theorem's hypotheses unverifiable and the proof's key line unjustified from the text. This does not change the reader's CONDITIONAL verdict; it reinforces it.","tokens_in":12585,"tokens_out":35903,"duration_ms":387135,"concrete_test":"Compute ∂∂ω^2 for the Hermitian metric ω_X = p_Y^*ω_Y + p_Z^*ω_Z in Example 3.3: verify that the only torsion contribution is p_Y^*(∂ω_Y∧∂̄ω_Y), which vanishes by type on the 2-dimensional Hopf surface factor. Also check that ∂ω∧∂ω is identically zero by bidegree for n≤3 but nonzero in general, confirming that the printed condition must be read as ∂ω∧∂̄ω=0.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Theorem 3.2 and Lemma 3.1 impose '∂∂ω = ∂ω∧∂ω = 0'. Since ∂ω has bidegree (2,1), the wedge ∂ω∧∂ω has bidegree (4,2), which cannot be the torsion term in the expansion ∂∂ω^{k+1} = ω∧∂∂ω^k + ω^k∧∂∂ω + 2k ω^{k-1}∧∂ω∧∂ω; the last term must be ∂ω∧∂̄ω (bidegree (3,3)). A reader following the printed hypothesis cannot verify Lemma 3.1 or the mass bound in Theorem 3.2. The mass bound's step 'By the assumption on ω and integrating by part' requires ddc(ω^{n-m-p})=0; under the corrected condition this holds, and the Leibniz-rule cross terms vanish because closed currents have vanishing dc. But as written, the condition is not well-formed, and the proof's key estimate is not checkable. This is a presentation error rather than a mathematical flaw, but it blocks verification of the central claim.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies the relative non-pluripolar product of closed positive currents on compact Hermitian manifolds. It claims that on compact complex manifolds equipped with a Hermitian form satisfying (as the authors intend) ∂∂ω=0 and ∂ω∧∂̄ω=0, the relative non-pluripolar product ⟨T1∧...∧Tm˙∧T⟩ is always well-defined (Theorem 3.2), and that the mass of such products is monotone under replacing each current by a less singular current in the same Bott-Chern class (Theorem 3.4). The main new ingredient is a global mass bound obtained by integrating by parts, using the condition that powers of ω are ∂∂-closed. A Hopf-surface times Kähler example is given to show the class is strictly larger than the Kähler class.","tokens_in":12771,"tokens_out":10513,"duration_ms":126317,"significance":"If the results are correct, they extend the theory of relative non-pluripolar products, previously developed on compact Kähler manifolds by Vu and others, to a family of non-Kähler Hermitian manifolds; this is a meaningful step because such products are used in complex Monge-Ampère theory and analytic intersection theory. The paper gives a self-contained treatment of several technical lemmas from Vu's paper and the core mass-bound argument is an independent contribution. However, as written, the central hypothesis is type-inconsistent and the key algebraic identity in Lemma 3.1 is incorrect, so the main theorem cannot be verified from the printed text. The issues appear fixable, but they are load-bearing.","major_comments":[{"comment":"The condition '∂∂ω = ∂ω ∧ ∂ω = 0' used throughout is type-inconsistent: ∂∂ω has bidegree (2,2), while ∂ω∧∂ω has bidegree (4,2). Consequently the expansion in the proof of Lemma 3.1, ∂∂ω^{k+1} = ω∧∂∂ω^k + ω^k∧∂∂ω + 2kω^{k-1}∧∂ω∧∂ω, is not of the correct bidegree; the last term must be ∂ω∧∂̄ω, which has bidegree (3,3) when wedged with ω^{k-1}. As printed, the equivalence (i)⇔(iii) in Lemma 3.1 is false, and the hypothesis of Theorem 3.2 does not imply ddc(ω^{n-m-p})=0, which is exactly what the mass-bound proof uses. The manuscript should consistently replace the condition by ∂∂ω = 0 and ∂ω∧∂̄ω = 0 and re-check the coefficient in the corrected expansion.","section":"Lemma 3.1, Theorem 3.2, Theorem 1.1"},{"comment":"The step 'By the assumption on ω and integrating by part' that yields ∫_X (ddc u~_{1,k}+θ1+Cω)∧...∧(ddc u~_{m,k}+θm+Cω)∧T∧ω^{n-m-p} = ∫_X (θ1+Cω)∧...∧(θm+Cω)∧T∧ω^{n-m-p} is not justified as written. It requires ddc(S∧ω^{N})=0 for every closed positive current S and N=n-m-p; this follows from the corrected condition ∂∂ω^N=0 together with dS=0 and d^cS=0, but not from the printed hypothesis. In addition, the functions u~_{j,k} are only bounded quasi-psh, so the integration by parts must be justified by a smoothing argument (for instance, the sequence u~^l_{j,k} already introduced in the proof). Please add a short justification.","section":"Theorem 3.2, proof of mass bound"},{"comment":"Lemma 3.5 is stated and proved for an open subset U ⊆ C^n, but Theorem 3.4 is on a compact Hermitian manifold. The proof applies Lemma 3.5 directly to conclude 'S ≥ ⟨T1∧...∧Tm˙∧T⟩' without explaining how the local semicontinuity inequality is globalized to the compact manifold. One needs to apply the lemma in coordinate charts and observe that the inequality between positive currents is local, or to prove a global version of the lemma. Without this patch, the monotonicity proof has a gap.","section":"Theorem 3.4, proof"}],"minor_comments":[{"comment":"The abstract contains a duplicated article: 'the the relative non-pluripolar product'.","section":"Abstract"},{"comment":"The statement 'endowed with a Hermitian form ω satisfied {θ1}∂∂,...,{θm}∂∂' is garbled; it should read 'satisfying one of the conditions in Lemma 3.1'.","section":"Theorem 1.3 and Theorem 3.4"},{"comment":"The statement contains the typos 'The the following reults hold' instead of 'The following results hold'.","section":"Lemma 2.4"},{"comment":"In the proof, the definition 'T'_{j,k}=ddc u'_j + θ_j' should be 'T'_{j,k}=ddc u'_{j,k} + θ_j'.","section":"Proposition 3.6"},{"comment":"The symbol 'ws.k' in the integration-by-parts argument should be 'w_{s,k}'.","section":"Proposition 3.6, proof"},{"comment":"In item (iv), 'T has non mass' should be 'T has no mass'; in item (v), the expression 'Tm ˙T' is missing a wedge sign, and the notation for the pole set of the difference current should be clarified.","section":"Proposition 2.7"}],"recommendation":"major_revision","confidential_remarks":"The paper imports the construction of relative non-pluripolar products and several technical lemmas from Vu's paper, so the genuinely new part is the mass estimate under the Hermitian condition. The type-inconsistent condition in Lemma 3.1 and Theorem 3.2 is likely a typo, but it must be fixed before the paper is publishable. The Hopf×Kähler example does verify the intended corrected condition because ∂ω_Y∧∂̄ω_Y vanishes automatically on a complex surface. I would also ask the authors to clarify how the local semicontinuity lemma is globalized in the proof of Theorem 3.4; the omission is probably routine, but it currently leaves a gap. The paper is within scope of the journal and, after these revisions, could be a useful contribution."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick take: this is a credible, useful extension of Vu's relative non-pluripolar product from Kähler to a class of Hermitian manifolds satisfying ∂∂ω=0 and ∂ω∧∂̄ω=0. The proof strategy is Vu's, adapted cleanly, and the mass bound is the key new step. The paper is worth refereeing, but the authors need to fix a genuine type error in the statement of the main condition.\n\nWhat's new: Theorem 3.2 proves well-definedness of ⟨T1∧...∧Tm∧̇T⟩ on compact Hermitian manifolds with that metric condition. That goes beyond the Kähler setting. The example (product of Hopf surface with Kähler) shows the class is non-empty. Theorem 3.4 gives mass monotonicity, a useful variant of Vu's cohomological monotonicity; since Bott-Chern cohomology lacks Poincaré duality, the mass statement is the right formulation.\n\nWhat's done well: the paper is honest about following Vu; the local-to-global machinery is laid out and the key mass estimate is genuinely independent. The adaptation of Lemma 3.5 and Proposition 3.6 follows the template but the exposition is reasonably careful.\n\nSoft spots: (1) The condition in Theorem 1.1, Lemma 3.1, and Theorem 3.2 is stated as ∂∂ω = ∂ω∧∂ω = 0. As written, ∂ω∧∂ω has bidegree (4,2), so it cannot be the cross term in ∂∂ω^{k+1}. The intended condition is ∂∂ω=0 and ∂ω∧∂̄ω=0. This is a presentation error, but it matters: the integration by parts in the mass bound is not checkable without it. Lemma 3.1's equivalence (i)-(iii) only works with the corrected term. (2) In the proof of Theorem 3.4, the use of the local semicontinuity Lemma 3.5 on the compact manifold is a bit quick; one needs to patch local pieces with a partition or use a global statement. It's likely fixable, but as written it's a gap. (3) There are typos (e.g., '∂∂ω=∂∂ω^2=0' in Lemma 3.1(ii) missing superscript, 'the the' in abstract, etc.) and the rendering of equations is sloppy.\n\nThe mathematics, modulo the typo, appears sound. No circularity, no parameter fitting. The class of Hermitian manifolds is a genuine restriction (Inoue surfaces don't admit such metrics), but it includes non-Kähler products, so the result is not vacuous.\n\nWho should read it: people working on pluripotential theory on Hermitian manifolds, or applications to complex Monge-Ampère equations. It deserves a serious referee and likely publication after minor revision.\n\nRecommendation: send it to review. The referee should ask for the corrected metric condition and a small clarification of the global step in Theorem 3.4.","headline":"Solid extension of Vu's relative non-pluripolar product to a restricted Hermitian class, but the main metric condition is misprinted as ∂ω∧∂ω instead of ∂ω∧∂̄ω.","tokens_in":13350,"tokens_out":2511,"would_cite":true,"duration_ms":28351,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["32U15","32U40"],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper proves that the relative non-pluripolar product of closed positive currents is well-defined on compact Hermitian manifolds whose Hermitian form $\\omega$ satisfies $\\partial\\bar\\partial\\omega=0$ and…","keywords":["relative non-pluripolar product","closed positive current","plurisubharmonic function","Monge-Ampère operator","compact Hermitian manifold","pluriclosed Hermitian metric","Bott-Chern cohomology"],"falsifier":"The central claim is settled by computing, on a compact Hermitian manifold whose metric does not satisfy the two equations, the truncated masses $\\|\\mathbf{1}_{\\cap\\{u_j>-k\\}}(dd^c u_{1,k}+\\theta_1)\\wedge\\cdots\\wedge(dd^c u_{m,k}+\\theta_m)\\wedge T\\|$ for closed positive currents with $m+p\\le n$; if these masses diverge as $k\\to\\infty$, the relative product is not well-defined and the metric condition in Theorem 3.2 is necessary as well as sufficient. A concrete search would start with a compact complex surface that admits no pluriclosed metric, since there the condition reduces to $\\partial\\bar\\partial\\omega=0$.","tokens_in":12341,"feed_emoji":"📐","tokens_out":15551,"duration_ms":166272,"temperature":0.7,"pith_summary":"This paper shows that the relative non-pluripolar product of closed positive currents is always defined on a class of compact Hermitian manifolds that properly contains the compact Kähler manifolds. The extra room comes from allowing a Hermitian metric $\\omega$ with $\\partial\\bar\\partial\\omega=0$ and $\\partial\\omega\\wedge\\bar\\partial\\omega=0$, instead of demanding Kählerity. On such manifolds the truncated approximants of the product have a mass bound that is independent of the truncation level, so the monotone limit that defines the product is guaranteed to exist. The same condition yields a monotonicity result: replacing a factor by a less singular current in the same Bott-Chern class cannot decrease the mass of the relative product.","feed_headline":"Relative current products exist beyond Kähler manifolds","feed_subtitle":"A Hermitian metric with ∂∂ω=0 and ∂ω∧∂̄ω=0 suffices for well-defined products and mass monotonicity.","key_machinery":"The central object is the relative non-pluripolar product $\\langle T_1\\wedge\\cdots\\wedge T_m\\,\\dot\\wedge\\,T\\rangle$, an intersection of closed positive $(1,1)$-currents with an arbitrary closed positive $(p,p)$-current, defined by truncating the potentials at level $k$ and taking the limit of the truncated wedges restricted to the set where all potentials exceed $-k$. The identity carrying the argument is Lemma 3.1: for a smooth $(1,1)$-form $\\omega$, the conditions $\\partial\\bar\\partial\\omega=0$ and $\\partial\\omega\\wedge\\bar\\partial\\omega=0$ are equivalent to $\\partial\\bar\\partial\\omega^k=0$ for every $k\\ge 1$. This is what makes the integration-by-parts mass estimate independent of the truncation level and therefore makes the limiting product well-defined.","core_discovery":"On a compact complex manifold $X$ of dimension $n$ with a Hermitian form $\\omega$ satisfying $\\partial\\bar\\partial\\omega=0$ and $\\partial\\omega\\wedge\\bar\\partial\\omega=0$, for any closed positive $(1,1)$-currents $T_1,\\dots,T_m$ and any closed positive $(p,p)$-current $T$ with $p+m\\le n$, the relative non-pluripolar product $\\langle T_1\\wedge\\cdots\\wedge T_m\\,\\dot\\wedge\\,T\\rangle$ is a well-defined positive current of finite mass. The proof represents each $T_j=dd^c u_j+\\theta_j$, truncates the potentials at level $k$, and shows that the corresponding truncated wedges have mass bounded by the $k$-independent integral $\\int_X(\\theta_1+C\\omega)\\wedge\\cdots\\wedge(\\theta_m+C\\omega)\\wedge T\\wedge\\omega^{n-m-p}$. The condition on $\\omega$ is exactly what lets every integration by parts eliminate the $dd^c$ factors, and Lemma 3.1 rewrites it equivalently as $\\partial\\bar\\partial\\omega^k=0$ for all $k\\ge 1$. For currents in the same Bott-Chern class, the paper also proves that replacing $T_j$ by a less singular $T_j'$ does not decrease the mass: $\\|\\langle T_1\\wedge\\cdots\\wedge T_m\\,\\dot\\wedge\\,T\\rangle\\|\\le\\|\\langle T_1'\\wedge\\cdots\\wedge T_m'\\,\\dot\\wedge\\,T\\rangle\\|$.","pith_inferences":["Inference: because the mass estimate is local, the same construction should go through on non-compact Hermitian manifolds admitting a complete metric satisfying the two equations, provided a global exhaustion by compact sets with controlled mass is available.","Inference: it remains open whether $\\partial\\bar\\partial\\omega=0$ alone (without $\\partial\\omega\\wedge\\bar\\partial\\omega=0$) is sufficient in higher dimensions; a natural test is a pluriclosed non-Kähler manifold of dimension at least three, where the extra term need not vanish by degree reasons.","Inference: a counterexample on a compact surface admitting no pluriclosed metric (such as an Inoue surface) would show the metric condition is not merely sufficient but necessary, and would pinpoint the obstruction to relative products in general Hermitian geometry.","Inference: the truncation-independent mass formula suggests interpreting $\\int_X(\\theta_1+C\\omega)\\wedge\\cdots\\wedge(\\theta_m+C\\omega)\\wedge T\\wedge\\omega^{n-m-p}$ as a Hermitian analogue of an intersection number, possibly leading to a Chern-class-valued intersection theory for currents on non-Kähler manifolds."],"forward_implications":["The theorem extends the known Kähler-case well-definedness of relative non-pluripolar products to every compact Hermitian manifold carrying a Hermitian form $\\omega$ with $\\partial\\bar\\partial\\omega=0$ and $\\partial\\omega\\wedge\\bar\\partial\\omega=0$.","In complex dimension two the condition reduces to $\\partial\\bar\\partial\\omega=0$, so the result covers every compact complex surface that admits a pluriclosed Hermitian metric.","The monotonicity statement holds for masses rather than cohomology classes, since Poincaré duality for Bott-Chern cohomology can fail on Hermitian manifolds.","Products formed with less singular representatives of the same Bott-Chern classes carry at least as much mass, which gives a comparison principle for the corresponding Monge-Ampère-type operators in this Hermitian setting.","The Hopf-surface product example supplies explicit non-Kähler manifolds to which both theorems apply, so the enlargement beyond Kähler manifolds is nonempty."],"supporting_citations":[{"why":"Supplies the definition of the relative non-pluripolar product, the truncation-level construction, and the local convergence lemmas and comparison results the proofs invoke.","marker":"[20]"},{"why":"Establishes the Kähler-case well-definedness that this paper extends, and provides the closedness of non-pluripolar products used in the theory.","marker":"[5]"},{"why":"Gives the regularization of quasi-plurisubharmonic functions on compact manifolds used to approximate truncated potentials by smooth ones before integrating by parts.","marker":"[4]"},{"why":"Provides the model non-Kähler Hermitian metrics on Hopf-surface products that show the class of manifolds covered by Theorem 3.2 is strictly larger than the Kähler class.","marker":"[19]"},{"why":"Lays the fine-topology and Monge-Ampère foundations for non-pluripolar products, used in the local comparison behind the construction.","marker":"[3]"}],"fun_headline_variants":["Relative non-pluripolar products exist on Hermitian manifolds","Mass monotonicity for relative current products","Beyond Kähler: relative current products well-defined","Hermitian manifolds admit relative product mass bounds","Relative current products: existence and monotonicity"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the Hermitian metric satisfies $\\partial\\bar\\partial\\omega=0$ and $\\partial\\omega\\wedge\\bar\\partial\\omega=0$; without this condition the truncated masses need not be bounded independently of the truncation level, and there are compact Hermitian manifolds on which no such metric exists.","fun_headline_variants_meta":{"raw":{"variants":["Relative non-pluripolar products exist on Hermitian manifolds","Mass monotonicity for relative current products","Beyond Kähler: relative current products well-defined","Hermitian manifolds admit relative product mass bounds","Relative current products: existence and monotonicity"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000352,"raw_usage":{"total_tokens":1903,"prompt_tokens":915,"completion_tokens":988,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":531,"completion_tokens_details":{"reasoning_tokens":914}},"tokens_in":531,"tokens_out":988,"duration_ms":10461,"temperature":1.0,"reasoning_tokens":914,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T12:18:23.592528+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"The central claim is settled by computing, on a compact Hermitian manifold whose metric does not satisfy the two equations, the truncated masses $\\|\\mathbf{1}_{\\cap\\{u_j>-k\\}}(dd^c u_{1,k}+\\theta_1)\\wedge\\cdots\\wedge(dd^c u_{m,k}+\\theta_m)\\wedge T\\|$ for closed positive currents with $m+p\\le n$; if these masses diverge as $k\\to\\infty$, the relative product is not well-defined and the metric condition in Theorem 3.2 is necessary as well as sufficient. A concrete search would start with a compact complex surface that admits no pluriclosed metric, since there the condition reduces to $\\partial\\bar\\partial\\omega=0$.","supporting_citations":[{"cited_title":"V U, Relative non-pluripolar product of currents , Ann","cited_arxiv_id":null,"evidence_quote":"Supplies the definition of the relative non-pluripolar product, the truncation-level construction, and the local convergence lemmas and comparison results the proofs invoke."},{"cited_title":"B OUCKSOM , P","cited_arxiv_id":null,"evidence_quote":"Establishes the Kähler-case well-definedness that this paper extends, and provides the closedness of non-pluripolar products used in the theory."},{"cited_title":"B Ł OCKI AND S","cited_arxiv_id":null,"evidence_quote":"Gives the regularization of quasi-plurisubharmonic functions on compact manifolds used to approximate truncated potentials by smooth ones before integrating by parts."},{"cited_title":"T OSATTI AND B","cited_arxiv_id":null,"evidence_quote":"Provides the model non-Kähler Hermitian metrics on Hopf-surface products that show the class of manifolds covered by Theorem 3.2 is strictly larger than the Kähler class."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Lays the fine-topology and Monge-Ampère foundations for non-pluripolar products, used in the local comparison behind the construction."}],"review_version":1}