{"id":"55905fd8-6541-45f8-9a07-822e4af94cca","arxiv_id":"2502.00623","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Compact Kähler manifolds with pseudo-effective tangent bundle admit a smooth fibration whose base is an étale quotient of a torus and whose fibers are rationally connected.","lead":"This paper proves that any compact Kähler manifold with a pseudo-effective tangent bundle can be sliced into a torus base and rationally connected fibers. It extends a structure theorem known for projective varieties to all Kähler manifolds, completing a program posed by Matsumura.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Theorem 2.3(2) hinges on [Wu22, Main Theorem] to conclude the subsheaf F of Ω_X is numerically flat, but the hypotheses of that theorem are not stated or verified, so the proof may fail exactly where X is assumed non-projective.","rationale":"The central claim of Theorem 1.1 rests on proving that a very general Albanese fiber F is projective and rationally connected. The key step is Theorem 2.3(2), which rules out non-projectivity of X when qhat(X)=0. The pivotal move inside that theorem is the conclusion that the reflexive sheaf F (the image of TX in Ω_X) is numerically flat. This conclusion is imported from [Wu22] and [MWb] without stating their exact hypotheses. Since the contradiction that follows (trivialization of F over a finite étale cover, yielding a non-zero 1-form) depends entirely on F being flat, an unverified or inapplicable hypothesis in [Wu22] would break the proof of projectivity and hence the rational connectivity of the fibers. This is more load-bearing than the extension of [Mul, Lemma 3.1] to non-projective bases, which affects only the local-constancy assertion (5) and is explicitly acknowledged with a proof sketch. The reader's weakest assumption identifies the same hinge, and I agree. A careful check of [Wu22]'s statement would settle the matter; if it applies at the stated Kähler generality, the proof is likely sound, and if not, the theorem is not proven.","tokens_in":7433,"tokens_out":27937,"duration_ms":272114,"concrete_test":"Read [Wu22, Main Theorem, Corollary] and list its hypotheses. Check (i) whether the theorem is stated for compact Kähler manifolds or requires projectivity; (ii) whether the F constructed in Theorem 2.3(2) (a strongly pseudo-effective reflexive subbundle of Ω_X with strongly pseudo-effective dual and c1(det F)=0) satisfies each hypothesis. If the theorem requires projectivity, the proof of Theorem 2.3(2) is invalid. If it does not, verify that the non-nef locus or any other condition holds for F; a failure of any single condition would leave the numerical flatness step unproved.","verdict_should_be":"UNCHANGED","load_bearing_attack":"In the proof of Theorem 2.3(2), after constructing F as the reflexive hull of the image of a non-zero holomorphic 2-form σ: TX → Ω_X, the authors assert that F is a subvector bundle of Ω_X and then conclude that F is numerically flat by applying [Wu22, Main Theorem, Corollary] (and cf. [MWb, Theorem 2.11]). The argument only verifies that F and F* are strongly pseudo-effective and that c1(det F)=0. It does not state what [Wu22, Main Theorem] requires. In particular, at this point X is assumed to be a compact Kähler manifold that is not projective; if [Wu22, Main Theorem] is stated only for projective varieties, the application is invalid and the contradiction that establishes projectivity of X collapses. Even if the theorem is Kähler-compatible, F must satisfy every additional hypothesis (e.g., a non-nef-locus condition, vanishing discriminant, or a semistability condition) that [Wu22] may impose. The paper's reliance on an external statement without reproducing it or checking its hypotheses is the weakest link in the chain from pseudo-effective tangent bundle to rational connectedness of the Albanese fibers.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proves Theorem 1.1: if X is a compact Kähler manifold whose tangent bundle is pseudo-effective (in the strong sense of admitting singular Hermitian metrics on symmetric powers with psh weights), then X admits a smooth fibration onto a finite étale quotient of a compact torus, with very general fibers rationally connected and having pseudo-effective tangent bundle; if the tangent bundle admits a positively curved singular Hermitian metric, the fibration is locally constant. The proof uses the Albanese map, reduces to the case of vanishing augmented irregularity, and then shows that such an X must be projective and rationally connected by combining Campana's theory of special varieties with a flat subbundle argument. The paper also derives the corollary that π1(X) is virtually abelian.","tokens_in":7681,"tokens_out":13989,"duration_ms":143329,"significance":"If the proof is correct, this is a substantial extension of the structure theorem for pseudo-effective tangent bundles from the projective case to all compact Kähler manifolds, resolving a problem posed in [Mat22b, Problem 4.5]. The paper is proof-theoretic, with no fitted parameters, and the final inductive structure of the argument is clear. Its main strength is the clean conceptual reduction: Theorem 2.2 establishes specialness (and hence virtual abelianity of linear representations of the fundamental group), and Theorem 2.3(2) converts the existence of a flat subbundle of the cotangent bundle into a contradiction when the augmented irregularity vanishes. However, the proof relies heavily on external results, and several load-bearing steps are cited rather than proved or even stated in sufficient detail; these gaps currently prevent full verification of the main theorem.","major_comments":[{"comment":"In the proof of Theorem 2.3(2), the assertion 'by applying [Wu22, Main Theorem, Corollary] to F, we conclude that F is a numerically flat locally free sheaf' is not supported by the text: the hypotheses of [Wu22, Main Theorem] are not stated, and the proof has not verified that they hold for the reflexive subsheaf F of Ω_X on a compact Kähler manifold (which may be non-projective). The flatness of F is load-bearing, as it produces the representation ρ whose virtual abelianity and finiteness yield the contradiction q(X')=0. Please state [Wu22, Main Theorem] in full and check each hypothesis, or provide a self-contained Kähler proof of the numerical flatness of F.","section":"Section 2, proof of Theorem 2.3(2)"},{"comment":"The proof uses the closure property that a generically surjective quotient of a pseudo-effective sheaf is pseudo-effective, first for (π∗L)∗ in Theorem 2.2 and then for F∗ in Theorem 2.3(2). This property is neither proved nor referenced in the manuscript; for the metric definition of pseudo-effectivity used here, it requires a separate argument (e.g., via regularized metrics or the non-nef locus description). Since this property is used to conclude that (π∗L)∗ has a Hermitian flat metric and that c1(det F)=0, a proof or precise reference is needed.","section":"Section 2, proof of Theorem 2.2 and Theorem 2.3(2)"},{"comment":"The removal of the projectivity assumption on Y from [Mul, Lemma 3.1] is only sketched. The sketch says that the first two steps of the proof in [Mul] do not require Y projective, while the third step uses [Mul, Theorem 1.6 and Proposition 1.7] and appeals to [Bis95, Remark 3.7.(ii)] and Simpson's correspondence for compact Kähler manifolds. This is not a complete proof of the Kähler version of the lemma. Because local constancy is one of the main conclusions of Theorem 1.1, the authors should either prove the Kähler version of [Mul, Lemma 3.1] or state and prove the corresponding Kähler analogues of [Mul, Theorem 1.6 and Proposition 1.7].","section":"Section 2, proof of Theorem 1.1(5)"},{"comment":"The conclusion 'the image of any GL-representation ρ : π1(X) → GL(r,C) is virtually abelian' is cited to [Cam04, Theorem 7.8] without stating the theorem. Since this conclusion is used in the proof of Theorem 2.3(2) to make Im(ρ) abelian after a finite étale cover, the authors should quote the precise statement and confirm that its hypotheses are satisfied for compact Kähler manifolds with pseudo-effective tangent bundle. If [Cam04, Theorem 7.8] is a standard result, a precise statement with a reference suffices.","section":"Section 2, Theorem 2.2"}],"minor_comments":[{"comment":"[MWa] is cited for the definition of locally constant fibration but is absent from the reference list; please add the full bibliographic entry.","section":"Section 1, Theorem 1.1(5)"},{"comment":"The entry [Mok92] appears twice in the reference list; remove the duplicate.","section":"References"},{"comment":"The reduction to a finite étale cover is stated without proof: 'It is sufficient to prove the conclusion after we replace X with a finite étale cover by the argument in [CH19] and [Hör07, Corollary 2.11] (see also the proof of [Mat, Theorem 1.1])'. Please explain how the fibration on the cover descends to X, or give a precise lemma reference.","section":"Section 2, proof of Theorem 1.1"},{"comment":"The statement 'the vector bundle (Λ^r Ω_X ⊗ det F^*)^* = Λ^r TX ⊗ det F is pseudo-effective' would benefit from a justification; please provide a reference or a one-line argument for pseudo-effectivity after twisting by a numerically trivial line bundle.","section":"Section 2, proof of Theorem 2.3(2)"},{"comment":"There are minor typographical issues, e.g., 'morpshim' for 'morphism' in the proof of Theorem 1.1(5) and an anomalous space in 'Consequently' in the introduction.","section":"Throughout"}],"recommendation":"major_revision","confidential_remarks":"The central argument depends on several preprints and external results ([HIM22], [IMZ], [Mat], [Mul], [CH], [MWa], [MWb], [Ou], [Wu22], [Cam04]). The gaps identified are all of the form 'state and verify the hypotheses of the cited theorem', rather than known counterexamples, so the paper may be salvageable with a substantial revision that makes the external inputs explicit and self-contained enough for verification. The editors may wish to ask the authors to provide a version where the statements of [Wu22, Main Theorem], [Mul, Lemma 3.1], and [Cam04, Theorem 7.8] are reproduced and their applicability to the Kähler setting is checked in detail."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"First thing you should know: this is the Kähler-case completion of the structure theorem for pseudo-effective tangent bundles, and it uses a genuinely different strategy from the projective proof. Instead of MRC fibrations, the authors use the Albanese map, prove the source is of special type, and then reduce to showing the general fiber is projective. That reduction is the real work, and Theorem 2.2's argument that the source must be special—via a curvature current supported on an exceptional divisor—is clean and convincing. The final induction is also clearly laid out. The paper earns credit for tackling a problem that was explicitly posed and for writing the proof in a mostly self-contained way given its citations.\n\nThe soft spots are all about verification of cited results. The load-bearing step is Theorem 2.3(2): from a non-zero holomorphic 2-form, they build a reflexive subsheaf F of Ω_X that is strongly pseudo-effective with vanishing determinant, and then conclude F is numerically flat by invoking [Wu22, Main Theorem]. The stress-test note is on target here—the paper does not state what [Wu22] requires, and the authors do not check hypotheses. At that point X is not assumed projective, and if [Wu22] is only proved for projective varieties, the contradiction establishing projectivity collapses. This is a genuine gap in presentation, not a manufactured one. A referee needs to look at [Wu22] and confirm the theorem holds for compact Kähler manifolds, or ask the authors to supply a proof.\n\nThe other soft spot is the removal of projectivity in the base for [Mul, Lemma 3.1], used for local constancy. The authors give a three-step sketch and cite [Bis95] and Simpson's correspondence. This is probably fine, but it is a sketch, and since the lemma is imported from a preprint, it deserves a careful check. There are also several other preprints in the chain ([Mat], [IMZ], [MWb]). That's more of a refereeing nuisance than a mathematical flaw—the cited results are about projective or intermediate cases, and the main theorem isn't circular.\n\nSo my overall read: the central theorem is very likely true, the proof is a coherent chain of established and near-established results, and the weak points are verifiable rather than conceptual. This deserves a serious referee. I'd recommend sending it to review with a note that the referee should verify the applicability of [Wu22] at the stated level of generality and ask for a full proof of the extended [Mul, Lemma 3.1].","headline":"The Kähler case of the structure theorem is very likely true and the proof is a coherent chain; the main gap is unverified hypotheses in a cited theorem at the projectivity hinge.","tokens_in":8234,"tokens_out":3441,"would_cite":true,"duration_ms":31655,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["32J25","14F35","58A30"],"pacs":[],"model":"deepseek-v4-flash","headline":"Compact Kähler manifolds with pseudo-effective tangent bundle decompose into a torus base and rationally connected fibers.","keywords":["pseudo-effective vector bundle","compact Kähler manifold","rationally connected fibration","augmented irregularity","special type","virtually abelian fundamental group","Albanese map","singular Hermitian metric"],"falsifier":"Find a compact Kähler manifold with pseudo-effective tangent bundle and vanishing augmented irregularity that is not projective; Theorem 2.3(2) asserts no such manifold exists, so one explicit example would refute the main theorem.","tokens_in":7219,"feed_emoji":"🧩","tokens_out":7278,"duration_ms":63939,"temperature":0.7,"pith_summary":"This paper proves that every compact Kähler manifold whose tangent bundle is pseudo-effective admits a smooth fibration onto a finite étale quotient of a compact complex torus, with a very general fiber that is rationally connected and again has pseudo-effective tangent bundle. In the presence of a positively curved singular Hermitian metric on the tangent bundle, the fibration is locally constant rather than merely locally trivial. The result carries the structure theorem for smooth projective varieties over to the full Kähler category, settling a problem left open in that earlier work. A direct consequence is that the fundamental group of such a manifold is virtually abelian.","feed_headline":"Pseudo-effective tangent bundle gives torus base plus rational fibers","feed_subtitle":"Structure theorem now covers compact Kähler manifolds; positive curvature gives a locally constant fibration.","key_machinery":"The central objects are the Albanese map and the augmented irregularity $\\hat q(X)$, the supremum of the irregularity over finite étale covers. The argument first establishes that a compact Kähler manifold with pseudo-effective tangent bundle is of special type, which implies that every linear representation of its fundamental group has virtually abelian image. The decisive step, Theorem 2.3(2), shows that when $\\hat q(X)=0$ such a manifold is projective and rationally connected: a non-projective manifold would admit a flat subbundle of the cotangent bundle, and the external numerical-flatness criterion plus virtual abelianity force a contradiction with $\\hat q(X)=0$. The fibration of Theorem 1.1 is then built by applying the induction hypothesis to the fibers of the Albanese map.","core_discovery":"The main theorem asserts that for a compact Kähler manifold $X$ with pseudo-effective tangent bundle, the Albanese map $X \\to Y$ is a smooth fibration whose base $Y$ is a finite étale quotient of a torus, whose very general fiber $F$ is rationally connected and has pseudo-effective tangent bundle, and whose fibration becomes locally constant once $TX$ admits a positively curved singular Hermitian metric. The proof proceeds by showing that a fiber with vanishing augmented irregularity must be projective, using the fact that such an $X$ is of special type and hence has linear fundamental group representations with virtually abelian image. A contradiction argument then forces projectivity, and the projective structure theorem supplies rational connectivity. The smoothness of the fibration and the properties of the fibers are propagated by induction on dimension.","pith_inferences":["It would be natural to test whether, without positive curvature, the fibration of Theorem 1.1 still coincides with the MRC fibration; the paper shows this only in the positively curved case.","The mechanism suggests that pseudo-effectivity of the tangent bundle is strong enough to force the base of the Albanese map to be a torus quotient in all Kähler dimensions, which would rule out any non-toral base behavior.","One could attempt to make the projectivity step self-contained by proving the numerical flatness of cotangent subsheaves and the virtual abelianity directly in the Kähler category, removing the reliance on external theorems."],"forward_implications":["Every compact Kähler manifold with pseudo-effective tangent bundle has virtually abelian fundamental group (Corollary 1.3).","The Albanese map is a smooth fibration onto a finite étale quotient of a torus, so the non-rational part of the manifold is entirely accounted for by the base.","In the positively curved case the fibration is locally constant, giving a stronger rigidity than local triviality.","The structure theorem for smooth projective varieties is recovered as the projective case, with the same fiberwise conclusion."],"supporting_citations":[{"why":"Supplies the projective structure theorem used as the base case and Theorem 3.12, which shows the tangent bundle of a fiber of the Albanese map remains pseudo-effective.","marker":"[HIM22]"},{"why":"Provides the criterion that a strongly pseudo-effective reflexive subsheaf of the cotangent bundle is numerically flat, the key external input in Theorem 2.3(2).","marker":"[Wu22]"},{"why":"Defines special type and supplies the theorem used to conclude that linear representations of the fundamental group have virtually abelian image.","marker":"[Cam04]"},{"why":"Gives Proposition 3.12 for comparing augmented irregularities under étale covers and the lemma used to identify a flat cotangent subbundle.","marker":"[DPS94]"},{"why":"Provides Lemma 3.1, which the paper adapts to show the fibration is locally constant in the positively curved case.","marker":"[Mul]"},{"why":"Contributes the special-type strategy for compact Kähler manifolds that the proof of Theorem 2.2 follows.","marker":"[Mat]"}],"fun_headline_variants":["Pseudo-effective tangent bundle forces torus base with rational fibers","Torus fibration with rationally connected fibers from pseudo-effective tangent bundle","Pseudo-effective tangent bundle: torus base, rational fibers","Kähler manifolds with pseudo-effective tangent bundle are torus fibrations"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The proof relies on two external theorems: one asserting that a strongly pseudo-effective reflexive subsheaf of the cotangent bundle is numerically flat, and one asserting that special-type Kähler manifolds have linear fundamental group representations with virtually abelian image; if either theorem carries hidden hypotheses, the step showing that a fiber with vanishing augmented irregularity is projective breaks down.","fun_headline_variants_meta":{"raw":{"variants":["Pseudo-effective tangent bundle forces torus base with rational fibers","Torus fibration with rationally connected fibers from pseudo-effective tangent bundle","Pseudo-effective tangent bundle: torus base, rational fibers","Kähler manifolds with pseudo-effective tangent bundle are torus fibrations"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001791,"raw_usage":{"total_tokens":6967,"prompt_tokens":768,"completion_tokens":6199,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":384,"completion_tokens_details":{"reasoning_tokens":6124}},"tokens_in":384,"tokens_out":6199,"duration_ms":41001,"temperature":1.0,"reasoning_tokens":6124,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-09T18:17:34.548027+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Find a compact Kähler manifold with pseudo-effective tangent bundle and vanishing augmented irregularity that is not projective; Theorem 2.3(2) asserts no such manifold exists, so one explicit example would refute the main theorem.","supporting_citations":[],"review_version":1}