{"id":"e625df03-4a71-4dac-906f-2e8e5e297454","arxiv_id":"2610.37468","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":8.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"For every neutral polynomial, Siegel disk linearization holds if and only if the rotation number is Brjuno, resolving Douady's conjecture from the 1980s.","lead":"Complex-dynamics researchers prove that a polynomial with a neutral fixed point has a Siegel disk exactly when the rotation number meets the Brjuno condition, for every polynomial degree. The result closes a 40-year-old Douady conjecture and supplies uniform renormalization bounds for multicritical maps.","discovery_kind":"first_principles","skeptic_critique":{"model":"deepseek/deepseek-v4-flash","headline":"Theorem I is unconditional, but its only bridge from Zakeri-type polynomials to all neutral polynomials is the equality (1.19) quoted from the unpublished preprint [BCR26]; until that reduction is verified or published, the headline claim is conditional.","rationale":"The reader's weakest_assumption identifies the same load-bearing dependency: the reduction (1.19) from [BCR26] is the only bridge from the Zakeri case to Theorem I for all neutral polynomials, and [BCR26] is unpublished. I considered whether the more serious internal issue is the compressed final step of Theorem 8.19 or the dependence on [DL24] in Theorem 7.1. For the EGM rotation numbers actually used in the proof of Theorem I, the Brjuno function is finite, and the iterative selection of N′ in §8.4 decreases the level index at each step, so the termination argument is shorter but less obviously fatal. The [DL24] dependence supports the sectorial renormalization construction, but the paper supplies a substantial part of that argument in §7, whereas (1.19) is used verbatim and without proof. The Zakeri-level results (Theorems A, B, and D for EGM maps) are substantial and mostly proved in text, so the appropriate outcome is not rejection but a conditional verdict pending verification or publication of the [BCR26] reduction.","tokens_in":74012,"tokens_out":17253,"duration_ms":175395,"concrete_test":"Obtain the complete proof of [BCR26, Main Theorem] and verify that (1.19) is proved for every bounded-type θ and every g ≥ 2; then re-check the specific application in §1.6: for a non-Brjuno θ* with EGM approximants θ*,n, the equality must yield sup_{P_g(θ*,n)} ρ ≤ sup_{Zak_g(θ*,n)} ρ for the algebraic approximants f*,n (same coefficients, varied multiplier), with no hidden hypothesis on critical orbits, symmetry, or rotation-number growth. A referee-level check of that one reduction is the decisive test; if the preprint appears unchanged in a refereed venue, the concern is resolved.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is Theorem I. Its proof in §1.6 runs as follows: suppose a non-Brjuno θ* has a linearizable f* ∈ P_g(θ*); take EGM approximants θ*,n; Stability of rotational domains gives liminf ρ(f*,n) ≥ ρ(f*) > 0; Theorem D (1.18) bounds the conformal radius over Zak_g(θ*,n) by exp(−C_g B(θ*,n)); and the [BCR26] equality (1.19) is then needed to transfer that bound from the Zakeri locus to the actual algebraic approximants f*,n ∈ P_g(θ*,n). Thus (1.19) is not a peripheral citation: it is the step that upgrades the paper's Zakeri theorem to Douady's conjecture for all polynomials. The authors are transparent about this (Remark E and the proof of Theorem I modulo Theorem D), and [BCR26] is an arXiv preprint (2606.24833), not yet refereed or published. The concern is therefore not that a published result was cited instead of reproved; it is that an unpublished result is load-bearing in the statement of an unconditional theorem. A second unpublished dependence, [DL24], underlies the sectorial renormalization tower proof (Theorem 7.1), but (1.19) is the more decisive dependency for Theorem I itself. If [BCR26]'s Main Theorem has additional hypotheses, or if the equality (1.19) is proved only for a subclass such as real-symmetric polynomials, high-type rotations, or a restricted degree range, then the contradiction in §1.6 does not follow and Theorem I is not established by this paper.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek/deepseek-v4-flash","summary":"This paper develops a multicritical extension of the pseudo-Siegel renormalization theory introduced by Dudko and Lyubich for quadratic polynomials. For Zakeri-type polynomials, i.e., monic polynomials of degree g whose Siegel disk boundary contains all critical points, the authors construct almost-invariant multicritical pseudo-Siegel disks bZ^m (Theorems 6.1 and 6.7), establish uniform a priori bounds and a uniform quasiconformal bound for the top disk bZ^{-1}, and use these bounds to build sectorial renormalization towers (Theorem 7.1) with a compact tower attractor. They then prove an antirenormalization estimate (Lemma 8.1, announced as Lemma C), which is iterated to obtain Theorem D: for f in Zak_g(Theta_egm), -log rho_f is comparable to the Brjuno function B(theta), with comparison constants depending only on the degree. Combined with the Buff-Cheritat-Roesch reduction (1.19) from the preprint [BCR26], this is used in Section 1.6 to prove Theorem I, the optimality and sufficiency of the Brjuno condition for all neutral polynomials. The paper also states and partially addresses Conjectures II and III under the Zakeri restriction.","tokens_in":74605,"tokens_out":10811,"duration_ms":73818,"significance":"If the proof is valid, this is a major advance: it extends uniform pseudo-Siegel bounds from the unicritical case to multicritical Zakeri polynomials, provides the first quantitative multicritical neutral-renormalization estimate of the form (1.18), and, conditional on the [BCR26] reduction, resolves Douady's 1980s conjecture on the optimality of the Brjuno condition for polynomial Siegel disks. The paper is genuinely substantial: many of the load-bearing lemmas are proved in the text (the Log-Rule in parabolic fjords, the Snake, Welding, Calibration and Amplification lemmas, the Localization Lemma, the antirenormalization coordinate estimates), and the comparison constants in the main quantitative claims are declared to depend only on the degree. The compactness theorem for sectorial renorm-towers and the explicit finite-constant statements are strong, checkable contributions. The main reservations concern external dependencies: the reduction (1.19) is quoted from the unpublished preprint [BCR26], and the proof of Theorem 7.1 is delegated in part to the unpublished manuscript [DL24].","major_comments":[{"comment":"The equality sup_{f in P_g(theta)} rho(f) = sup_{f in Zak_g(theta)} rho(f) is quoted from the unpublished arXiv preprint [BCR26]. This is not a peripheral citation: it is exactly the step that transfers the Theorem D bound from the Zakeri class to the approximating maps f_{*,n} in the contradiction argument. Without this equality, the proof only establishes optimality of the Brjuno condition under the Zakeri restriction. The authors are transparent in Remark E and in the heading 'Consequence of Theorem D and [BCR26, Main Theorem]', but the abstract and Section 1.1 present Theorem I as a main outcome of the current paper. Because [BCR26] is not yet refereed or published, the headline theorem is conditional. The revision should either cite a refereed version, state Theorem I explicitly as conditional on the verification of (1.19), or prove the needed reduction. If (1.19) in [BCR26] has additional hypotheses, such as real symmetry, high type, or a restricted degree range, the contradiction in Section 1.6 does not go through.","section":"Section 1.6, Eq. (1.19)","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"Theorem B is load-bearing for Lemma 8.1, Proposition 7.8, and hence Theorem D, but its proof relies on the unpublished manuscript [DL24]. The sentence 'The proof follows the quadratic case [DL24], with appropriate adjustments to the multicritical setting' is not a proof of the multicritical adaptations, especially since the authors state in Section 1.7 that the absence of the No-Spiraling Property prevents them from directly employing various arguments of [DL24], and that only iterative antirenormalization changes in a small neighborhood of 0 are justified at this stage. As it stands, the sectorial renorm-tower statement, including Items (A)-(D) and the compactness in (7.3)-(7.4), is not fully exhibited in this text. This is a completeness issue rather than a demand to reproduce published work; [DL24] is a manuscript. The paper should list precisely which results are quoted from [DL24], which results are adapted, and what the adaptation proves in the multicritical case.","section":"Section 1.7 and Theorem 7.1","authors_acknowledge":"The parabolic geometry is more complicated when several critical orbits are involved; this is a key technical subtlety preventing us from directly employing various arguments of [DL24].","authors_acknowledge_p":0.7},{"comment":"The domain of Theorem D is ambiguous. In (1.5), Zak_g is defined as a union over bounded-type rotation numbers, and (1.7) then defines Zak_g(theta) for non-bounded theta as Zak_g cap P_g(theta), an intersection that is empty under (1.5). Theorem 8.19 states the result for f in Zak_g with Brjuno theta, which under (1.5) means only bounded type, while the proof first treats EGM angles and then uses stability of the Siegel disk to pass to Brjuno approximants. Please reconcile the definition of the Zakeri class for non-bounded theta with the compactification in (1.8), and state precisely over which theta-classes Theorem D is asserted.","section":"Sections 1.3 and 8.4","authors_acknowledge":null,"authors_acknowledge_p":null}],"minor_comments":[{"comment":"The same symbol Theta appears to be used for the set of irrationals and for its compactification; the typesetting of the second Theta is lost. Please disambiguate the two objects.","section":"Section 1.3","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"Lemma C is labeled '(Corollary 8.15)', but Corollary 8.15 proves the estimate (8.16) for the gluing map psi_anti, whereas the displayed estimate (1.16) is Lemma 8.1. This cross-reference should be corrected.","section":"Section 1.6, Lemma C","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"Item (C) says 'bZ^{-1} subset D contains a definite subdisk of D'; the notation bZ^{-1} should be bZ^{-1}_{f_n}, as the statement otherwise refers to an object that has not been introduced for the renormalized map.","section":"Theorem 7.1, Item (C)","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The displayed inequality involving |psi(w)-psi(-w)-2psi'(0)w| appears to have a typographical error: the factor '3-|z|^2/(1-|z|^2)^2' should presumably be '(3-|z|^2)/(1-|z|^2)^2', and the variable z should be related to w.","section":"Section 8.3, Eq. (8.15)","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"Theorem A is stated for theta in Theta_egm, while Theorem 6.1 is stated 'for all f in Zak_g(theta)' without the EGM restriction, and the last paragraph of Section 1.4 concludes Conjecture II for Zak_g(Theta). Please state explicitly over which theta-classes each theorem is proved.","section":"Section 1.4 and Theorem 6.1","authors_acknowledge":null,"authors_acknowledge_p":null}],"recommendation":"major_revision"},"author_rebuttal":null,"desk_editor":{"model":"deepseek/deepseek-v4-flash","letter":"You should know two things. First, the Zakeri-type result is real: Theorems A and D establish uniform multicritical pseudo-Siegel bounds and the comparison -log rho_f ~ B(theta) with constants depending only on the degree, for all neutral polynomials whose Siegel disk absorbs all critical points. That is a first-principles generalization of the quadratic framework to arbitrary degree with several critical orbits, and the proof is organized as an explicit chain (Theorems 6.1, 6.7, 7.1, Lemma 8.1, Theorem 8.19) with many lemmas proved in the text. The shortcut metric, the fjord-by-fjord construction without tunnels, and the sectorial renorm-tower compactification are genuinely new machinery. There are no fitted parameters; B(theta) is defined from the continued fraction alone, and the comparison constants depend only on the degree. The authors also explicitly flag what comes from elsewhere: Remark E and the proof of Theorem I modulo Theorem D quote the reduction (1.19) from the arXiv preprint [BCR26], and Theorem 7.1 follows the unpublished manuscript [DL24]. That transparency is creditable.\n\nSecond, the soft spots are exactly the dependencies. The equality (1.19)--sup over all bounded-type polynomials equals sup over the Zakeri locus--is load-bearing for Theorem I, not peripheral. It is the only bridge from the Zakeri theorem to Douady's conjecture for all neutral polynomials. Since [BCR26] is unreviewed and unpublished, Theorem I is conditional on it, even though the paper states it as a theorem. If (1.19) carries extra hypotheses, the contradiction in section 1.6 does not go through. The [DL24] dependence is less decisive for the headline but still real: Theorem 7.1 says the proof follows the quadratic case with adjustments, so the inductive core of sectorial renormalization is not fully exhibited here. A minor third point: the final step of Theorem 8.19, the termination of the Brjuno blow-up cascade, is compressed into a sketched repetition argument rather than a fully closing argument. That one is minor relative to the scale of the paper, but it is where a referee should ask for detail.\n\nFor the right reader--a holomorphic dynamics specialist--this is serious work deserving a full referee report. I would accept for peer review and ask the referee to verify (1.19) against [BCR26] and to check the termination argument in Theorem 8.19. My own advice: engage with it, cite it for the Zakeri results, and treat the full Douady claim as conditional until [BCR26] is published.","headline":"A genuine and substantial proof of the Zakeri-case Brjuno optimality, with the full Douady conjecture resting on an unreviewed arXiv reduction that the authors transparently quote.","tokens_in":75799,"tokens_out":752,"duration_ms":35658},"pith_extraction":{"msc":["37F50","37F25","37F10"],"pacs":[],"model":"deepseek/deepseek-v4-flash","headline":"A neutral polynomial has a linearizable Siegel disk if and only if its rotation number is Brjuno — the paper settles the 1980s conjecture via uniform multicritical pseudo-Siegel bounds.","keywords":["Siegel disks","Brjuno condition","neutral polynomials","pseudo-Siegel bounds","sectorial renormalization","Cremer hedgehog","conformal radius","small periodic points"],"falsifier":"Numerically compute, in the cubic slice $P_3(\\theta)$ with $\\theta$ the golden mean, the maximal conformal radius over the whole parameter slice and the radius attained on the Zakeri arc $Zak_3(\\theta)$: any strict surplus of the slice maximum over the Zakeri value falsifies the reduction (1.19) and with it Theorem I. Separately, for eventually golden-mean approximations $\\theta_n \\to \\theta_*$ with $B(\\theta_*) = \\infty$, check that $-\\log \\rho(f_n)/B(\\theta_n)$ for Zakeri $f_n$ stays within a degree-dependent band; a ratio tending to $0$ or $\\infty$ would falsify Theorem D's comparability, while a bounded ratio supports it.","tokens_in":75491,"feed_emoji":"🌀","tokens_out":15567,"duration_ms":119447,"temperature":0.7,"pith_summary":"This paper aims to settle a conjecture from the 1980s: a polynomial with an irrationally neutral fixed point has a linearizable Siegel disk around that point if and only if its rotation number satisfies the Brjuno condition, the same arithmetic criterion that governs analytic circle diffeomorphisms. The paper's own contribution is the Zakeri case: whenever the boundary of the Siegel disk carries all critical points, the conformal radius $\\rho_f$ of the maximal linearization domain satisfies $-\\log \\rho_f \\asymp B(\\theta)$ with comparison constants depending only on the degree. The route runs through new multicritical pseudo-Siegel a priori bounds (Theorem A), a compact family of sectorial renormalization towers (Theorem B), and an antirenormalization estimate (Lemma C) showing that the Brjuno summations themselves locate the small periodic points that obstruct linearization. Fed through the reduction of [BCR26] — that the largest bounded-type Siegel disks are attained by Zakeri polynomials — the comparison yields Theorem I for all neutral polynomials. If correct, the result extends the quadratic theorem of [Y95] to every degree and shows that the non-linearizable pathologies of local germs cannot occur in polynomial dynamics.","feed_headline":"Brjuno condition proved optimal for all neutral polynomials","feed_subtitle":"New pseudo-Siegel bounds settle the 1980s conjecture: Siegel disks exist exactly for Brjuno rotation numbers.","key_machinery":"The load-bearing objects are the pseudo-Siegel disks: a nested family $\\widehat Z_{-1} \\supset \\widehat Z_0 \\supset \\cdots \\supset Z_f$ of almost-invariant topological disks enveloping the Siegel disk, constructed level by level by filling in parabolic fjords so that the geometric irregularities of the boundary are absorbed while uniform bounds are preserved. Because several critical points sit on $\\partial Z_f$, one combinatorial level may contain several orbits of preparabolic intervals at different depths; the paper replaces the quadratic-case tunnels with a shortcut metric in which every dam has length $l_{m+1}$, and the Snake, Welding, Calibration, and Amplification Lemmas convert wide submerging laminations into outer degeneration on deeper scales (Theorem A). On top of these bounds, sectorial renormalization towers $(f_n : \\mathbb{D} \\dashrightarrow \\mathbb{D})_{n\\ge 0}$ (Theorem B) form a precompact, shift-invariant attractor whose parabolic limits have multiplicity at most $g$. The final mechanism is the antirenormalization coordinate $w \\mapsto \\hat\\theta_n w + b(-\\log\\hat\\theta_n) + O(1)$ near the neutral point (Lemma C): its iterates reproduce the Brjuno summations, and every large partial sum predicts a small periodic point that blocks linearization.","core_discovery":"Stated on the paper's own terms, the central result is Theorem D: for every Zakeri polynomial of a fixed degree $g$, the conformal radius $\\rho_f$ of the maximal linearization domain around the neutral fixed point obeys $-\\log \\rho_f \\asymp B(\\theta)$, where $B(\\theta) = \\sum_n \\log(q_{n+1}/q_n)$ is the Brjuno function and the comparison constants depend only on $g$. The proof first treats eventually golden-mean rotation numbers, where the sectorial renormalization tower is explicit, and then extends to all Zakeri polynomials by approximating arbitrary $\\theta$ by such numbers and invoking stability of the rotational domain [ABC20]. The mechanism is arithmetic: the antirenormalization map near the origin behaves like $w \\mapsto \\hat\\theta_n w - \\log \\hat\\theta_n$ in the logarithmic coordinate, so iterating it reproduces the Brjuno sums; whenever a partial sum is large, a small fixed point of a renormalized map lifts to a small periodic point of the original polynomial, and a periodic point cannot lie inside the Siegel disk. Thus a non-Brjuno rotation number forces the linearization domain to have radius at most $e^{-cB(\\theta)}$, and as $\\theta$ approaches non-Brjuno values the Siegel disk degenerates into the Cremer hedgehog. With the [BCR26] reduction this becomes Theorem I: for every polynomial of degree $\\ge 2$ with an irrationally neutral fixed point, linearizability is equivalent to the Brjuno condition.","pith_inferences":["If the reduction of [BCR26] is sound, the quantitative comparability $-\\log \\rho_f \\asymp B(\\theta)$ should extend from Zakeri maps to all neutral polynomials; Remark E of the paper points toward the sharper bound (1.20) with factor $1/(g-1)$, which would pin down the exact rate at which Siegel disks collapse as $\\theta$ becomes non-Brjuno.","The affine model $w \\mapsto \\hat\\theta_n w - \\log \\hat\\theta_n$ is an iterated map whose orbits encode $B(\\theta)$; one could treat the whole renormalization tower as a shadow of this model and obtain location estimates for small periodic points in degree $g \\ge 3$ that bypass the full a priori machinery.","A directly testable consequence: in the cubic slices $P_3(\\theta)$ of bounded type, the maximal Siegel radius over the slice should coincide with the radius on the Zakeri arc; any measurable surplus would locate the failure in the reduction step rather than in the pseudo-Siegel bounds.","The authors acknowledge that the absence of a No-Spiraling property in the multicritical case confines their antirenormalization estimates to a definite small neighborhood of $0$; extending that control is the bottleneck to the advertised rigidity program, in which Zakeri loci $Zak_3(\\theta)$ would become parameter Jordan arcs depending continuously on $\\theta$."],"forward_implications":["For every polynomial of degree $\\ge 2$ with an irrationally neutral fixed point, the Siegel disk exists if and only if the rotation number is Brjuno, closing the gap left open since the quadratic case [Y95] and ruling out local-germ pathologies in polynomial dynamics.","The size of the maximal linearization domain is arithmetically controlled: $-\\log \\rho_f$ grows like the Brjuno function, so the disk shrinks at a rate set by the continued-fraction denominators $q_n$.","As rotation numbers approach non-Brjuno values, the Siegel disk does not vanish abruptly but degrades into the Cremer hedgehog, and the Mother Hedgehog conjecture — a maximal rotating continuum whose boundary absorbs a critical orbit — holds for all Zakeri polynomials.","The Brjuno partial sums acquire a dynamical reading: each large sum predicts the location of a small periodic point, giving a constructive obstruction to linearization rather than a pure nonexistence argument.","The compact renorm-attractor of Theorem B lays groundwork for a non-perturbative neutral renormalization theory in degrees $\\ge 3$, with parabolic limits of bounded multiplicity."],"supporting_citations":[{"why":"supplies the reduction (1.19) that the proof of Theorem I invokes verbatim: for bounded-type rotation numbers, the supremum of Siegel-disk radii over all neutral polynomials equals the supremum over Zakeri polynomials.","marker":"[BCR26]"},{"why":"establishes the quadratic case of Theorem I and provides the Brjuno summations, the local sector renormalization framework, and the lower bound $-\\log \\rho \\succeq B(\\theta)$ used in Theorem D.","marker":"[Y95]"},{"why":"is the unicritical source that Theorem A generalizes; the Log-Rule, Snake, Calibration, and Amplification lemmas are adapted from it to the multicritical setting.","marker":"[DL22]"},{"why":"supplies the stability of rotational domains (Lemma 17) used to pass from eventually golden-mean rotation numbers to arbitrary ones and to obtain one half of the comparability in Theorem I.","marker":"[ABC20]"},{"why":"is the unpublished quadratic treatment of sectorial renormalization towers that the proof of Theorem B says it follows with multicritical adjustments.","marker":"[DL24]"},{"why":"formulates the conjecture, resolved here as Theorem I, that the Brjuno condition characterizes linearizability of neutral polynomial fixed points.","marker":"[Do87]"},{"why":"with [Rüs67], supplies the sufficiency half of Theorem I: Brjuno rotation numbers guarantee analytic linearization of the germ.","marker":"[Brj71/72]"}],"fun_headline_variants":["Brjuno condition optimal for all neutral polynomials","Douady conjecture resolved: Brjuno iff linearizable","Siegel disks exist exactly for Brjuno rotations","Multicritical bounds prove Brjuno optimality","Neutral polynomials: Brjuno condition is optimal"],"cache_read_input_tokens":0,"weakest_assumption_plain":"The headline theorem rests on an unpublished reduction the paper quotes from [BCR26]: for rotation numbers whose continued-fraction entries are bounded, the largest possible Siegel disk among all neutral polynomials of a given rotation number is the same as the largest among Zakeri polynomials, whose disk boundary carries every critical point; the paper's own renormalization-tower construction likewise follows an unpublished quadratic-case manuscript [DL24], so if either of those supports gives way, Theorem I does not follow even though the Zakeri bounds are the paper's own work.","fun_headline_variants_meta":{"raw":{"variants":["Brjuno condition optimal for all neutral polynomials","Douady conjecture resolved: Brjuno iff linearizable","Siegel disks exist exactly for Brjuno rotations","Multicritical bounds prove Brjuno optimality","Neutral polynomials: Brjuno condition is optimal"]},"model":"deepseek/deepseek-v4-flash","effort":"low","cost_usd":4.5e-05,"raw_usage":{"cost":4.46446e-05,"is_byok":false,"cost_details":{"upstream_inference_cost":4.46446e-05,"upstream_inference_prompt_cost":2.71726e-05,"upstream_inference_completions_cost":1.7472e-05},"total_tokens":1161,"prompt_tokens":1065,"completion_tokens":96,"prompt_tokens_details":{"audio_tokens":0,"video_tokens":0,"cached_tokens":958,"cache_write_tokens":0},"completion_tokens_details":{"audio_tokens":0,"image_tokens":0,"reasoning_tokens":33}},"tokens_in":1065,"tokens_out":96,"cost_source":"openrouter_usage","duration_ms":3625,"image_pages":[],"temperature":1.0,"billed_attempts":1,"reasoning_tokens":33,"transport_provider":"openrouter","cache_read_input_tokens":0,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-10-28T05:55:53.390280+00:00","model_set":{"reader":"deepseek/deepseek-v4-flash"},"pipeline_version":"pith-pipeline@v1.4.1-daily-deepseek-or-medium","input_kind":null,"input_chars":null,"input_total_chars":null,"passes_missing":null,"published_standing":null,"published_confidence":null,"objection_audit":null,"run_record":null,"basis":{"kind":"not_recorded","source":null,"chars_given":null,"chars_total":null},"receipt":null,"falsifier":"Numerically compute, in the cubic slice $P_3(\\theta)$ with $\\theta$ the golden mean, the maximal conformal radius over the whole parameter slice and the radius attained on the Zakeri arc $Zak_3(\\theta)$: any strict surplus of the slice maximum over the Zakeri value falsifies the reduction (1.19) and with it Theorem I. Separately, for eventually golden-mean approximations $\\theta_n \\to \\theta_*$ with $B(\\theta_*) = \\infty$, check that $-\\log \\rho(f_n)/B(\\theta_n)$ for Zakeri $f_n$ stays within a degree-dependent band; a ratio tending to $0$ or $\\infty$ would falsify Theorem D's comparability, while a bounded ratio supports it.","supporting_citations":[],"lane_accuracy":{"status":"not yet measured","lane":{"model":"deepseek/deepseek-v4-flash","input_kind":null},"url":"https://pith.science/accuracy"},"review_version":1}