{"id":"abce7cf2-ddb3-4d78-a564-0da8cd330802","arxiv_id":"2607.02815","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.5,"correctness_risk":"low","formal_verification":"none","parameter_count":0,"one_line_summary":"Suitably refined noncuspidal Saito-Kurokawa points for GSp4 are p-adically rigid under Kisin and monodromy hypotheses, so any interpolating family is a point.","lead":"Certain refined noncuspidal Saito-Kurokawa representations of GSp4 cannot be interpolated by nontrivial positive-dimensional p-adic families. This gives a GSp4 analogue of Bellaïche rigidity and isolates an obstruction to variation on the GSp4 eigenvariety.","discovery_kind":"extension","skeptic_critique":{"model":"grok-4.5","headline":"No significant objection identified beyond the reader's already-flagged (SK-P2) hypothesis.","rationale":"The central claim (Theorem 8.4) is a clean conditional rigidity statement whose logical chain closes under the listed hypotheses. The only place the implication can fail is precisely the monodromy-control condition already identified by the reader; the Kisin property is likewise standard and expected. Arthur's classification is used only for the discrete spectrum of GSp4 and is now essentially unconditional. No internal inconsistency, hidden circularity, or additional soft assumption was found. The verdict therefore remains CONDITIONAL with high confidence, and no adjustment is required.","tokens_in":32186,"tokens_out":559,"duration_ms":5224,"concrete_test":"On any concrete eigenvariety component known to contain a noncuspidal ψ_{2}- or ψ_{3}-refined SK point (e.g., the Siegel eigenvariety of Boxer–Pilloni or Berger–Betina), verify that the monodromy operators of classical points of general type satisfy (SK-P2)(i)–(ii) by comparing local Langlands parameters at primes of Steinberg type for μ; if the monodromy jumps outside the Zariski closure of Nπ,ℓ on a Zariski-dense set, the hypothesis fails for that component and Theorem 8.4 does not apply.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The reader's weakest-assumption diagnosis is accurate and already isolates the single softest link: condition (SK-P2) (Definition 8.2) is imposed rather than proved for the abstract families of Definition 5.4. Without it, the monodromy operators Nℓ(ρK) need not vanish or stay inside the conjugacy class of Nπ,ℓ, so Propositions 8.8–8.9 fail to force the Ext classes into the Bloch–Kato Selmer groups that vanish by Lemma 8.1. The rest of the argument (SK rigidity in §6, non-(2,1,1)-reducibility of Tη in Proposition 7.10, GMA extensions, Kisin crystallinity, and Selmer vanishing under (St) and noncuspidality) is internally consistent and correctly adapted from Bellaïche–Chenevier. No further load-bearing gap appears once (SK-P2) and the Kisin property are granted.","agreement_with_reader":"agree"},"referee_report":{"model":"grok-4.5","summary":"The paper proves p-adic rigidity for certain noncuspidal Saito–Kurokawa (SK) points on GSp4: a noncuspidal ψi-refined SK point z0 (i∈{2,3}, with a slope condition when i=3) that satisfies the Atkin–Lehner condition (St) cannot lie on a positive-dimensional irreducible p-adic family X(1) of Galois type with trivial central character, provided the family satisfies the monodromy-control condition (SK–P2) and the Kisin property at z0 (Theorem 8.4). The argument proceeds by abstracting the expected properties of a p-adic family (Definition 5.4), proving an auxiliary SK rigidity result for families with a dense set of SK points (§6), showing that the generic pseudocharacter Tη cannot be (2,1,1)-reducible (Proposition 7.10), and then using GMA machinery to produce a crystalline, unramified-outside-p extension that yields a nontrivial class in a vanishing Bloch–Kato Selmer group H1f(Q,ρμ(2)) or H1f(Q,Qp(1)).","tokens_in":32465,"tokens_out":1193,"duration_ms":9087,"significance":"The result is a genuine GSp4 analogue of Bellaïche’s rigidity theorems for U(2,1) and supplies a concrete obstruction to p-adic variation of noncuspidal SK lifts on the GSp4 eigenvariety. The adaptation of the Bellaïche–Chenevier GMA/pseudocharacter framework, the careful analysis of accessible refinements (Proposition 3.3 and Table 6.1), and the clean reduction to known Selmer vanishings (Kato, Kummer) are technically solid. The paper correctly isolates the role of the refinement (ψ2/ψ3 versus ψ1/ψ4) and of noncuspidality, and it makes the dependence on the monodromy hypothesis (SK–P2) and the Kisin property fully explicit. These are valuable contributions to the geometry of noncuspidal loci on higher-rank eigenvarieties.","major_comments":[{"comment":"Definition 8.2 and Theorem 8.4: the monodromy-control condition (SK–P2) is imposed as a hypothesis on the abstract families of Definition 5.4 rather than deduced from them. Propositions 8.8–8.9 rely on it to force ExtT classes into the Bloch–Kato Selmer groups that vanish by Lemma 8.1. Without (SK–P2) the contradiction does not close. The manuscript should either (a) prove that any family satisfying Definition 5.4 automatically satisfies (SK–P2) under the standing assumptions, or (b) restate the main theorem as a conditional rigidity statement for families that satisfy (SK–P2) and the Kisin property, and discuss the extent to which known eigenvariety constructions are expected to obey it (cf. Remark 8.3).","section":null},{"comment":"§2.3 and the proof of Proposition 7.10: Arthur’s classification is assumed (with a reference to recent progress on the twisted weighted fundamental lemma). The partition of Z into Zef ⊔ ZSK and the claim that type-(a) points are irreducible and type-(b)/(c) points are precisely (2,2)-reducible are load-bearing for the non-(2,1,1)-reducibility of Tη. A short, self-contained statement of precisely which parts of Arthur’s classification are used, and a clear indication of the residual dependence on the fundamental lemma, would make the logical status of Proposition 7.10 transparent.","section":null}],"minor_comments":[{"comment":"Definition 5.4: the notion of an abstract p-adic family is carefully adapted from Bellaïche, but a brief comparison with the properties known to hold on existing GSp4 eigenvarieties (e.g., those of Pilloni, Boxer–Pilloni, or Andreatta–Iovita–Pilloni) would help the reader assess how restrictive the definition is.","section":null},{"comment":"Table (6.1) and Remark 6.2: the labelling of refinements when v(αz)=k(z)−3/2 is declared immaterial, but a one-line verification that the valuations of F1 and F2 still produce the same contradictions in Propositions 6.3–6.5 would remove any residual ambiguity.","section":null},{"comment":"Lemma 2.3 and Remark 2.4: the restriction to F=Q is correctly motivated by the vanishing of H1f(Q,Qp(1)), but a sentence on whether the argument could be adapted to totally real fields under additional assumptions on units would be useful for future work.","section":null},{"comment":"Typographical: the arXiv identifier appears as 2607.02815; confirm consistency of numbering and cross-references (e.g., “Proposition 6.3” vs. “Prop. 6.3”) throughout.","section":null}],"recommendation":"major_revision","confidential_remarks":"The paper is a solid, carefully written contribution that correctly identifies its own softest hypothesis. Once (SK–P2) is either proved or cleanly packaged as a standing assumption, the manuscript should be suitable for a strong number-theory journal. The dependence on Arthur is standard in the field and does not, in my view, warrant rejection."},"author_rebuttal":null,"desk_editor":{"model":"grok-4.5","letter":"This paper gives a clean GSp4 version of Bellaïche’s rigidity for certain noncuspidal Saito–Kurokawa points. The main theorem (8.4) says that a ψ2- or ψ3-refined noncuspidal SK point satisfying (St) and a mild slope condition cannot sit in a positive-dimensional p-adic family that obeys the Kisin property and monodromy control (SK-P2). That is new; the literature has deformations for holomorphic/cuspidal SK points and for ordinary refinements, but not this rigidity statement.\n\nWhat works well is the adaptation of the Bellaïche–Chenevier GMA/pseudocharacter machine. The local refinement table (6.1), the auxiliary SK-family rigidity in §6, and the non-(2,1,1) argument for the generic pseudocharacter (Prop. 7.10) are carefully done. Once you grant Kisin crystallinity and (SK-P2), the Ext classes land in H1f(Q, ρμ(2)) or H1f(Q, Qp(1)), both of which vanish (Kato + Kummer). The contradiction is clean and non-circular. Citations are appropriate; Arthur is taken from recent work that is now essentially unconditional for these purposes.\n\nThe soft spot is exactly the one the reader flagged: (SK-P2) is imposed on the abstract families of Definition 5.4 rather than proved. Without it the monodromy operators need not stay small enough for the Ext classes to be unramified outside p, so the Selmer vanishing does not close. Kisin is likewise a hypothesis, though expected on any actual eigenvariety. These are real limitations for an unconditional statement about the GSp4 eigenvariety itself, but they are clearly labelled and do not break the internal logic.\n\nThis is for people working on eigenvarieties, CAP representations, or p-adic families for GSp4. It is not a broad reorganisation of the subject, but it is a genuine new theorem that a serious referee should see. I would send it out.","headline":"Solid GSp4 analogue of Bellaïche rigidity for noncuspidal SK points; the argument holds under the stated hypotheses, with (SK-P2) the only real soft link.","tokens_in":33078,"tokens_out":536,"would_cite":true,"duration_ms":6885,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["11F33","11F46","11F80","11R39"],"pacs":[],"model":"grok-4.5","headline":"Certain refined noncuspidal Saito–Kurokawa points on GSp4 cannot sit in nontrivial p-adic families.","keywords":["p-adic rigidity","GSp4","Saito–Kurokawa","eigenvariety","pseudocharacters","Bloch–Kato Selmer groups","refinements","GMA"],"falsifier":"Exhibit a positive-dimensional irreducible component of a GSp4 eigenvariety (or an abstract p-adic family satisfying the paper’s axioms) that passes through a noncuspidal ψ2- or ψ3-refined SK point and still obeys both (SK–P2) and the Kisin property; such a component would refute the theorem.","tokens_in":33046,"feed_emoji":"🔒","tokens_out":783,"duration_ms":8287,"temperature":0.7,"pith_summary":"The paper shows that suitably refined noncuspidal automorphic Saito–Kurokawa representations of GSp4 over the adeles of Q are p-adically rigid: they do not deform in any positive-dimensional p-adic family that satisfies two natural interpolation conditions. The argument is an analogue of Bellaïche’s rigidity theorems for U(2,1). After an auxiliary rigidity result that already rules out pure SK families for the “middle” refinements, the author attaches a Galois pseudocharacter to a hypothetical family and shows that its generic fibre cannot be (2,1,1)-reducible. The residual GMA structure then produces a nontrivial crystalline extension class in one of two Bloch–Kato Selmer groups that both vanish. The vanishing of those Selmer groups is the obstruction that forces the family to be zero-dimensional. The result therefore isolates a concrete geometric obstruction on the GSp4 eigenvariety and explains why these particular noncuspidal points remain isolated.","feed_headline":"Noncuspidal Saito–Kurokawa points stay isolated p-adically","feed_subtitle":"Middle refinements of GSp4 SK lifts cannot deform once monodromy and crystallinity are controlled","key_machinery":"The GMA structure of the residual Cayley–Hamilton algebra of the family pseudocharacter at z0, which realises extension classes in ExtT(ϵ−2,ϵ−1) or ExtT(ϵ−2,ρµ). Combined with the Kisin property (crystallinity at p) and (SK–P2) (unramified outside p), these classes land in Bloch–Kato Selmer groups that vanish, yielding the contradiction.","core_discovery":"Theorem 8.4 asserts that a noncuspidal ψi-refined Saito–Kurokawa point z0 (i = 2 or 3, with a mild slope condition when i = 3) that satisfies a local Atkin–Lehner sign condition (St) cannot lie on any irreducible positive-dimensional p-adic family of Galois-type automorphic representations that obeys the monodromy-control condition (SK–P2) and the Kisin property at z0; any such family is necessarily a single point.","pith_inferences":[],"forward_implications":[],"fun_headline_variants":["Noncuspidal GSp4 SK points stay rigid under monodromy control","Refined Saito–Kurokawa lifts of GSp4 cannot form p-adic families","Isolated noncuspidal SK points on the GSp4 eigenvariety","Middle-refined SK representations of GSp4 resist p-adic variation","GSp4 SK points satisfying (St) remain single points in families"],"cache_read_input_tokens":14464,"weakest_assumption_plain":"The monodromy-control condition that every classical point of the family has Galois monodromy at primes away from p no larger than that of the original Saito–Kurokawa representation; if monodromy can jump, the unramifiedness needed for Selmer vanishing fails.","fun_headline_variants_meta":{"raw":{"variants":["Noncuspidal GSp4 SK points stay rigid under monodromy control","Refined Saito–Kurokawa lifts of GSp4 cannot form p-adic families","Isolated noncuspidal SK points on the GSp4 eigenvariety","Middle-refined SK representations of GSp4 resist p-adic variation","GSp4 SK points satisfying (St) remain single points in families"]},"model":"grok-4.5","effort":"low","cost_usd":0.005094,"raw_usage":{"total_tokens":1373,"prompt_tokens":690,"num_sources_used":0,"completion_tokens":105,"cost_in_usd_ticks":50940000,"prompt_tokens_details":{"text_tokens":690,"audio_tokens":0,"image_tokens":0,"cached_tokens":256},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":578,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":690,"tokens_out":105,"duration_ms":5582,"temperature":1.0,"reasoning_tokens":578,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-07-12T06:51:25.262742+00:00","model_set":{"reader":"grok-4.5"},"falsifier":"Exhibit a positive-dimensional irreducible component of a GSp4 eigenvariety (or an abstract p-adic family satisfying the paper’s axioms) that passes through a noncuspidal ψ2- or ψ3-refined SK point and still obeys both (SK–P2) and the Kisin property; such a component would refute the theorem.","supporting_citations":[],"review_version":1}