{"id":"86a8c8f3-719f-453d-9c9c-89d74e9cb52b","arxiv_id":"2211.06776","paper_version":4,"verdict":"UNVERDICTED","confidence":"LOW","novelty_score":6.0,"correctness_risk":"unknown","formal_verification":"none","parameter_count":0,"one_line_summary":"Extends the LLV algebra to primitive symplectic varieties with isolated singularities via an isomorphism g ≅ so((IH²(X,Q), Q_X) ⊕ h) and studies the resulting representation theory with applications to the P=W conjecture.","lead":"The paper proves an isomorphism for the total Lie algebra acting on the intersection cohomology of primitive symplectic varieties that have only isolated singularities. A generalist might care because it supplies a purely algebraic route to results that previously needed analytic tools and offers applications to a conjecture linking different cohomology filtrations.","discovery_kind":"extension","skeptic_critique":{"model":"grok-4.3","headline":"Extension of LLV algebra to IH* assumes hard Lefschetz and representation properties hold verbatim despite isolated singularities","rationale":"The reader's weakest assumption correctly isolates the single point where the argument is least secure: the verbatim transfer of the smooth-case structural theorems to intersection cohomology. No other internal inconsistency is visible from the claim and abstract; the algebraic proof for the smooth case is a side benefit but does not remove the need to justify the singular extension.","tokens_in":1720,"tokens_out":439,"duration_ms":15072,"concrete_test":"Locate the section defining the LLV algebra g on IH^* and the proof that it is isomorphic to so((IH^2,Q_X)⊕h). Extract the step that verifies the hard Lefschetz property or the sl(2)-action on the Verbitsky component for singular X; if that step only cites the smooth-case theorems without an additional argument for isolated singularities, recompute the dimension of the degree-2 part of the Verbitsky component on a concrete example (e.g., a K3 surface with an isolated A_1 singularity resolved by a small resolution) and check whether it matches the expected dimension from the smooth formula.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim requires that the Lie algebra generated by the Lefschetz operators (from the symplectic class and the intersection BBF form Q_X) on IH^*(X,Q) satisfies exactly the same relations and yields the same so((V ⊕ h)) structure as in the smooth case of Looijenga-Lunts-Verbitsky. For X with isolated singularities this is not automatic: intersection cohomology satisfies Poincaré duality and a Hodge structure, but the hard Lefschetz theorem, the primitive decomposition, and the fact that the Verbitsky component is irreducible under the LLV action must be re-established or shown to follow from the smooth locus. The paper's abstract invokes the extension directly; if the proof only reduces to the smooth case via resolution or stratification without controlling the contribution of the singular points to the cup-product and the resulting Lie bracket, the isomorphism may fail to hold.","agreement_with_reader":"agree"},"referee_report":{"model":"grok-4.3","summary":"The paper extends results of Looijenga-Lunts and Verbitsky to show that the total Lie algebra g acting on the intersection cohomology of a primitive symplectic variety X with isolated singularities satisfies g ≅ so((IH²(X, Q), Q_X) ⊕ h), where Q_X is the intersection Beauville-Bogomolov-Fujiki form and h is a hyperbolic plane. It studies IH^*(X, Q) as a g-representation with emphasis on the Verbitsky component, multidimensional Kuga-Satake constructions and Mumford-Tate algebras, and gives applications to the P=W conjecture. The approach yields a new algebraic proof of the smooth case that avoids the hyperkähler metric.","tokens_in":1889,"tokens_out":633,"duration_ms":23342,"significance":"If the isomorphism holds, the result is significant: it generalizes the LLV algebra beyond the smooth setting and supplies an algebraic derivation for irreducible holomorphic symplectic manifolds. The explicit study of the Verbitsky component and Mumford-Tate algebras under the g-action, together with the applications to P=W, are concrete strengths. The algebraic route is a clear advantage over metric-dependent arguments.","major_comments":[{"comment":"§4, Theorem 4.1 and the surrounding derivation of the Lie bracket relations: the claim that the LLV algebra on IH^* is identical to the smooth case requires an explicit verification that the hard Lefschetz theorem, primitive decomposition, and irreducibility of the Verbitsky component continue to hold when isolated singularities are present; the contribution of the singular locus to the cup-product structure that defines the Lie brackets is not controlled in the given argument, which is load-bearing for the stated isomorphism.","section":"§4"},{"comment":"§5.3, the multidimensional Kuga-Satake construction: the extension of the representation-theoretic statements from the smooth case is invoked without a separate check that the Hodge structure on IH^* remains of the expected weight and that the Mumford-Tate algebra commutes with the LLV action in the singular setting; this step is used to derive the applications to P=W and therefore needs direct justification.","section":"§5.3"}],"minor_comments":[{"comment":"Notation for the intersection BBF form Q_X is introduced without an explicit comparison to the usual BBF form on the smooth locus; a short remark clarifying the relation would improve readability.","section":"§2"},{"comment":"The abstract states the main result but does not indicate the key technical step (control of cup products away from the singularities) that distinguishes the argument from a direct citation of Looijenga-Lunts-Verbitsky.","section":"Abstract"}],"recommendation":"major_revision","confidential_remarks":"The manuscript cites the foundational LLV papers appropriately, but the novelty claim for the singular case would benefit from a clearer statement of how much of the proof is genuinely new versus a reduction to the smooth locus."},"author_rebuttal":{"model":"grok-4.3","summary":"We thank the referee for the careful reading and the detailed report. The two major comments identify places where the manuscript would benefit from additional explicit checks when extending from the smooth to the isolated-singularities setting. We address each point below and will incorporate the requested verifications in the revised version.","responses":[{"response":"The Lie-algebra generators and their bracket relations are defined entirely in terms of the intersection-cohomology ring structure on IH^*(X,Q). Because the singularities are isolated, the intersection product that enters the definition of the operators is computed on the smooth locus in all degrees relevant to the LLV algebra; the singular locus lies in codimension at least 2 and therefore does not contribute to the cup-product pairings that appear in the bracket relations. The hard Lefschetz theorem and the resulting primitive decomposition for IH^* with respect to an ample class are known to hold for varieties with isolated singularities. The irreducibility of the Verbitsky component then follows from the same representation-theoretic argument used in the smooth case, once the generators act on the same graded vector space. Nevertheless, we agree that spelling out these facts explicitly strengthens the exposition. In the revision we will insert a short paragraph (or lemma) in §4 that records the cited properties of intersection cohomology and confirms that the singular locus does not alter the Lie brackets.","revision_made":"yes","referee_comment":"[§4] §4, Theorem 4.1 and the surrounding derivation of the Lie bracket relations: the claim that the LLV algebra on IH^* is identical to the smooth case requires an explicit verification that the hard Lefschetz theorem, primitive decomposition, and irreducibility of the Verbitsky component continue to hold when isolated singularities are present; the contribution of the singular locus to the cup-product structure that defines the Lie brackets is not controlled in the given argument, which is load-bearing for the stated isomorphism."},{"response":"Intersection cohomology of a projective variety with isolated singularities carries a pure Hodge structure of the expected weight. The LLV operators are realized by cup-product with classes of type (1,1) and therefore preserve the Hodge filtration; consequently the Mumford-Tate algebra, which is generated by the Hodge classes, commutes with the LLV action by the same algebraic reason as in the smooth case. The multidimensional Kuga-Satake construction is then obtained verbatim from the representation theory of the LLV algebra. While these facts are standard, we acknowledge that a direct sentence or two confirming them in the singular setting would make the passage to the P=W applications fully self-contained. We will add this short justification to §5.3 in the revision.","revision_made":"yes","referee_comment":"[§5.3] §5.3, the multidimensional Kuga-Satake construction: the extension of the representation-theoretic statements from the smooth case is invoked without a separate check that the Hodge structure on IH^* remains of the expected weight and that the Mumford-Tate algebra commutes with the LLV action in the singular setting; this step is used to derive the applications to P=W and therefore needs direct justification."}],"tokens_in":1442,"tokens_out":670,"duration_ms":23912,"standing_objections":[]},"desk_editor":{"model":"grok-4.3","letter":"The main takeaway is that this work supplies a metric-free algebraic proof of the LLV isomorphism for smooth irreducible holomorphic symplectic manifolds and claims the same so((IH²(X,Q), Q_X) ⊕ h) structure for the intersection cohomology when X has only isolated singularities. Both the proof method and the singular extension look new relative to the cited Looijenga-Lunts and Verbitsky papers. It also works out the g-module structure on IH*, with focus on the Verbitsky component, Kuga-Satake constructions, and Mumford-Tate algebras, and flags some direct consequences for the P=W conjecture. That is useful bookkeeping for people already inside this corner of hyperkähler geometry. The soft spot is exactly the one raised in the stress-test note. Intersection cohomology satisfies Poincaré duality and carries a Hodge structure, but hard Lefschetz, the primitive decomposition, and the irreducibility of the Verbitsky summand under the LLV action are not automatic once singularities are present. The abstract states the isomorphism as if these properties transfer directly, so the body must contain a clear argument that the isolated points do not alter the cup-product relations or the generated Lie algebra. Without seeing explicit control of the local contributions at the singularities, it is hard to judge whether the claim holds or whether the proof reduces to the smooth locus too quickly. This is specialized reading for people already working on LLV algebras, intersection cohomology of symplectic varieties, or the P=W conjecture. It is not broad enough to interest a general algebraic geometer. The paper shows clear engagement with the literature and formulates a definite statement, so it is coherent on its own terms even if one disagrees with the extension. A serious editor should send it to referees who can check the singular-case arguments in detail.","headline":"The paper gives an algebraic proof of the LLV algebra for smooth IHS manifolds and extends the isomorphism to IH* of isolated-singularity cases, but the extension rests on unverified carry-over of hard Lefschetz and irreducibility.","tokens_in":2365,"tokens_out":452,"would_cite":false,"duration_ms":14560,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":{"model":"grok-4.3","evidence":[],"headline":"LLV algebra on intersection cohomology of singular symplectic varieties has no structural overlap with RS distinction-forcing chain","alignment":"orthogonal","rationale":"The paper's core results (Theorems 1.1, 5.4, 5.9) extend Looijenga-Lunts-Verbitsky Lie algebra structure to IH^*(X,Q) via Hard Lefschetz operators, symplectic symmetry (sl2×sl2 from σ,σ̄), monodromy density (Bakker-Lehn), and BBF form Q_X, yielding g ≅ so((IH²,Q_X)⊕h). This is classical algebraic geometry/Hodge theory with no reference to recognition cost J, golden-ratio ladders, 8-tick periodicity, or parameter-free derivation of constants. RS theorems (reality_from_one_distinction, AbsoluteFloorClosure, AlexanderDuality_circle_linking forcing D=3, Jcost uniqueness) operate on a completely different surface; the paper neither invokes nor contradicts any RS structural theorem.","tokens_in":64701,"confidence":"high","tokens_out":224,"duration_ms":7770,"cache_read_input_tokens":38528,"cache_creation_input_tokens":0},"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"grok-4.3","headline":"The total Lie algebra on intersection cohomology of a primitive symplectic variety with isolated singularities is the special orthogonal Lie algebra of the second cohomology plus a hyperbolic plane.","keywords":["primitive symplectic varieties","intersection cohomology","LLV algebra","Beauville-Bogomolov-Fujiki form","P=W conjecture","irreducible holomorphic symplectic manifolds","Lie algebra representations"],"falsifier":"An explicit computation of the LLV algebra for a concrete primitive symplectic variety with isolated singularities (for example a quotient singularity or a small resolution) whose dimension or bracket relations fail to match those of so((IH²(X, Q), Q_X) ⊕ h).","tokens_in":2596,"feed_emoji":"","tokens_out":725,"duration_ms":17117,"temperature":0.7,"pith_summary":"The paper extends the structural theorems of Looijenga-Lunts and Verbitsky to prove that the total Lie algebra g acting on the intersection cohomology of a primitive symplectic variety X with isolated singularities satisfies g ≅ so((IH²(X, Q), Q_X) ⊕ h). This supplies an algebraic proof of the corresponding statement for smooth irreducible holomorphic symplectic manifolds that does not use the hyperkähler metric. A sympathetic reader would care because the isomorphism organizes the entire graded intersection cohomology as a representation of a classical Lie algebra and supplies new algebraic tools for questions such as the P = W conjecture.","feed_headline":"Lie algebra on intersection cohomology equals so of BBF form plus hyperbolic plane","feed_subtitle":"The result extends to singular cases and supplies a metric-free algebraic proof for the smooth irreducible holomorphic symplectic setting.","key_machinery":"The LLV algebra (the total Lie algebra g generated by Lefschetz operators and their duals on intersection cohomology), shown to be isomorphic to the indicated special orthogonal Lie algebra.","core_discovery":"We extend results of Looijenga--Lunts and Verbitsky and show that the total Lie algebra g for the intersection cohomology of a primitive symplectic variety X with isolated singularities is isomorphic to g ≅ so((IH²(X, Q), Q_X) ⊕ h), where Q_X is the intersection Beauville--Bogomolov--Fujiki form and h is a hyperbolic plane. This gives a new, algebraic proof for irreducible holomorphic symplectic manifolds which does not rely on the hyperkähler metric.","pith_inferences":["If the same structural extension holds for varieties with non-isolated singularities, the LLV algebra description would apply more broadly.","The metric-free proof opens the possibility of comparing LLV algebras across birational models or deformations that change the singularity type.","Representation-theoretic invariants of g may produce new Hodge-theoretic constraints on the possible intersection cohomology rings."],"forward_implications":["The graded intersection cohomology IH^*(X, Q) becomes a representation of this orthogonal Lie algebra, with the Verbitsky component appearing as a distinguished summand.","Multidimensional Kuga-Satake constructions and Mumford-Tate algebras can be read off from the representation theory of g.","The algebraic description yields immediate consequences for the P = W conjecture on primitive symplectic varieties.","The same isomorphism holds for smooth irreducible holomorphic symplectic manifolds via a purely algebraic argument."],"fun_headline_variants":["LLV algebra equals so of BBF form plus hyperbolic plane","LLV algebra for primitive symplectic varieties equals so of BBF form plus hyperbolic plane","LLV algebra of intersection cohomology is so of BBF form plus hyperbolic plane","LLV algebra for singular symplectic varieties is so of BBF form plus hyperbolic plane"],"cache_read_input_tokens":2112,"weakest_assumption_plain":"The structural results of Looijenga-Lunts and Verbitsky on the LLV algebra extend verbatim to the intersection cohomology of primitive symplectic varieties that possess only isolated singularities.","fun_headline_variants_meta":{"raw":{"variants":["LLV algebra equals so of BBF form plus hyperbolic plane","LLV algebra for primitive symplectic varieties equals so of BBF form plus hyperbolic plane","LLV algebra of intersection cohomology is so of BBF form plus hyperbolic plane","LLV algebra for singular symplectic varieties is so of BBF form plus hyperbolic plane"]},"model":"grok-4.3","cost_usd":0.012086,"raw_usage":{"total_tokens":5269,"prompt_tokens":654,"num_sources_used":0,"completion_tokens":80,"cost_in_usd_ticks":120862000,"prompt_tokens_details":{"text_tokens":654,"audio_tokens":0,"image_tokens":0,"cached_tokens":256},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":4535,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":654,"tokens_out":80,"duration_ms":24047,"temperature":1.0,"reasoning_tokens":4535,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-05-24T10:44:20.923351+00:00","model_set":{"reader":"grok-4.3"},"falsifier":"An explicit computation of the LLV algebra for a concrete primitive symplectic variety with isolated singularities (for example a quotient singularity or a small resolution) whose dimension or bracket relations fail to match those of so((IH²(X, Q), Q_X) ⊕ h).","supporting_citations":[],"review_version":1}