{"id":"e570c4ec-bc37-41d8-9761-6b94f5a3df3e","arxiv_id":"1909.02009","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Applying the vacuum 2-3 rotation and then repeatedly eliminating the largest off-diagonal matrix element makes the effective Hamiltonian converge with an error order that grows like the Fibonacci sequence.","lead":"This paper shows that a sequence of simple matrix rotations can solve neutrino oscillations in matter with exponentially fast precision, where the error level follows the Fibonacci sequence. The optimal recipe is to apply the vacuum 2-3 rotation first, then always remove the largest off-diagonal term, giving a fast and compact calculation method.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Fibonacci claim rests on an unproved resonant cancellation: at the solar resonance the Sec. 2.2 order-counting breaks down and the rescue (Sec. 3.1, Eq. 22) is only numerically demonstrated.","rationale":"The reader's weakest-assumption analysis identifies the same load-bearing concern: the Sec. 2.2 Fibonacci order-counting assumes O(1) diagonal gaps, and at the solar resonance this assumption fails. The paper itself flags the special-sequence caveat in Sec. 2.2 and provides only Eq. 22 plus numerical figures in support of the resonant cancellation. My stress test confirms that this is the most central unprotected step: without the cancellation, the exponential Fibonacci convergence would not hold uniformly in the matter potential, which is exactly what the abstract and conclusion claim. I considered whether a different issue, such as the lack of a proof of global optimality of LODE versus all other pivot strategies, is more load-bearing; it is not, because even a non-optimal pivot choice that still gives Fibonacci convergence would preserve the main quantitative claim, whereas a failure at the solar resonance would invalidate the claim for an important physical regime. The numerical evidence is real and the benchmark in Table 2 is encouraging, but the absence of a derivation and of released code leaves the resonant cancellation as a genuine soft spot. I therefore agree with the reader's CONDITIONAL verdict and do not see reason to move it in either direction. An honest non-finding is not appropriate here because the paper's own caveat identifies the exact point where the proof is missing.","tokens_in":11471,"tokens_out":9509,"duration_ms":106139,"concrete_test":"Implement the LODE sequence exactly as specified: vacuum (2-3) rotation first, then repeatedly rotate on the largest off-diagonal element, using exact numerical diagonalization of the 3x3 Hamiltonian in Eq. 12. Sweep a grid of Ye rho E values spanning the solar resonance, including the exact resonance point and offsets of 1%, 0.1%, and 0.01% on both sides, for both normal and inverted ordering. After each of the first 1-7 rotations, compute M_N = max_{j>k} |2E (H1)_jk / Delta_lambda_jk| and the corresponding eigenvalue-error metric, and check whether log10(M_N) grows with the Fibonacci exponents F_{N+1} and 2F_{N+1} without plateaus. If M_N is undefined at exact resonance, resolve the degenerate subspace by diagonalizing it and document the resulting value. Repeat the sweep with theta13, theta12, and Delta_m21^2 varied within their current 3-sigma ranges.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim that the LODE strategy achieves Fibonacci convergence for all matter potentials depends on the Sec. 2.2 order-counting argument, which assumes that the diagonal gap of the pivot pair is O(1) in units of the small parameter epsilon, so that sin(alpha) inherits the order of the off-diagonal element. At the solar resonance, after the recommended vacuum (2-3) plus matter (1-3) rotations, the relevant diagonal gap vanishes exactly at a = Delta_m21^2 cos(2 theta12) / c13^2, so the rotation angle is not small and the generic Fibonacci recurrence is not applicable. The paper's rescue is the assertion in Sec. 2.2 (after Eq. 10) that a special sequence gives y ~ s13 epsilon and a numerator-denominator cancellation in the eigenvector corrections, quantified by Eq. 22 in Sec. 3.1. This is not turned into a proof that the metric in Eq. 27 remains O(epsilon^{F_N}) for all subsequent rotations and all parameter values; the numerical evidence in Fig. 2 and Table 2 covers a benchmark slightly above the solar resonance and a range of matter potentials, but no code is provided and the stability of the cancellation under parameter variations is not established. Furthermore, at exactly the solar resonance the denominator Delta_lambda in Eq. 27 vanishes, so the first-order eigenvector-correction metric is singular and the paper does not specify how the degenerate subspace is handled before the next rotation. If the cancellation is not stable, the claimed all-matter-potential Fibonacci rate fails precisely in the region where perturbative methods also need special treatment.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies a Jacobi-like rotation method for diagonalizing the 3-flavor neutrino Hamiltonian in matter, in which one successively applies 2x2 unitary rotations to reduce off-diagonal elements. It claims that, with a suitable pivot-selection strategy (largest off-diagonal element, LODE) and after an initial vacuum (2-3) rotation, the size of the off-diagonal part decreases with the number of rotations following the Fibonacci sequence, leading to exponential convergence. The authors also compare LODE with the largest-rotation-angle strategy (LROT), concluding that LODE is superior or equal at all matter potentials, and they provide numerical figures and tables over a range of neutrino energies. A generic order-counting argument in Sec. 2.2 is used to explain the Fibonacci rate, and the special behavior near the solar resonance is handled by a claimed numerator-denominator cancellation in the eigenvector-correction metric.","tokens_in":11739,"tokens_out":6528,"duration_ms":64606,"significance":"If the result holds, the paper offers a practical high-precision approximation scheme for neutrino oscillations with a simple, constant-cost iterative step and a striking exponential convergence rate. The order-counting derivation of the Fibonacci rate is elegant and, away from resonance, convincing; the LODE-versus-LROT comparison is a genuine practical insight. The paper does not fit any free parameter to the observed convergence rate: the rotation angles are algebraic functions of the Hamiltonian and the numerical inputs are standard global-fit neutrino parameters. The central weakness is that the all-matter-potential claim, including the solar-resonance region, rests on a cancellation that is numerically demonstrated for one benchmark rather than derived, and the exact-resonance point is not explicitly treated.","major_comments":[{"comment":"The generic Fibonacci order-counting argument assumes that the diagonal gap of the pivot pair is O(1) in units of the small parameter epsilon, so that sin(alpha) inherits the order of the off-diagonal element. At the solar resonance this hypothesis fails, as the paper acknowledges. The rescue is the statement, after Eq. (10) and again after Eq. (22), that a special sequence produces a cancellation between numerator and denominator in the eigenvector corrections. This cancellation is not derived: Eq. (22) is an asymptotic expansion at a approximately Delta_m21^2, and no explicit expression for the eigenvector-correction metric is given to show that the cancellation persists through all subsequent rotations. Since the abstract claims the Fibonacci rate without restricting the matter-potential range, this is a load-bearing gap in the proof of the central claim.","section":"Sec. 2.2, Eqs. (8)-(10) and Sec. 3.1, Eq. (22)"},{"comment":"At the exact solar resonance, a = Delta_m21^2 cos(2 theta12)/c13^2, the diagonal elements lambda_- and lambda_0 coincide, so the metric in Eq. (27), max|2E (H1)_jk / Delta_lambda_jk|, has a zero denominator for the (1-2) pair after the vacuum (2-3) and matter (1-3) rotations. The first-order perturbation theory used to define this metric is not valid at this point, and the paper does not specify how the degenerate subspace is handled before the next rotation is applied. The numerical figures may avoid the exact resonance point, but the text claims convergence 'at the solar resonance' and therefore needs an explicit treatment of the degenerate case.","section":"Sec. 3.1, Eqs. (19)-(22) and Sec. 3.3, Eq. (27)"},{"comment":"The numerical evidence for the resonance-region cancellation is limited to one benchmark (Ye rho E = 0.250 GeV g/cm^3) in Table 2 and to curves in Fig. 2 without numerical values exactly at the resonance or scans over the oscillation parameters that enter the cancellation (theta13, theta12, the mass ordering, and delta_CP). No code is provided. Given that the claimed all-matter-potential Fibonacci convergence depends on the stability of this cancellation, the paper should supply either an analytic proof or a more systematic numerical demonstration that includes the exact resonance point and variations of the relevant parameters.","section":"Sec. 3.3, Table 2 and Fig. 2"}],"minor_comments":[{"comment":"The line 'sin 2theta12 = 0.31, sin 2theta13 = 0.022, sin 2theta23 = 0.58' is presumably a typo for sin^2(theta12), sin^2(theta13), and sin^2(theta23); the numerical values are the global-fit values for the squared sines, not for sin(2theta).","section":"Sec. 3.3"},{"comment":"The expression 'cos(1 3 cos^{-1}(...))' appears to be a formatting error and should read 'cos(1/3 cos^{-1}(...))'.","section":"Introduction"},{"comment":"The matrices displayed after Eq. (18) and in Eq. (B.1) are hard to parse because the 2x2 block structure is not typeset clearly; please add explicit row and column indices.","section":"Eqs. (18) and (B.1)"},{"comment":"The metric in Eq. (27) is called the size of the first-order corrections to the eigenvectors, but the relationship between that metric and the usual first-order perturbation expression should be stated more explicitly, in particular the definition of Delta_lambda_jk and the role of the factor 2E.","section":"Sec. 3.3, Eq. (27)"}],"recommendation":"major_revision","confidential_remarks":"The paper is short and the main idea is attractive, but the unconditional abstract claim is not fully supported away from the generic O(1) gap regime. I would ask the authors to fix the resonance-region treatment and provide either an analytic argument or a more comprehensive numerical scan; this seems achievable within the scope of the manuscript."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Two things to know. First, this paper gives a clean and genuinely useful observation: the Jacobi-style rotation scheme for diagonalizing the 3x3 neutrino Hamiltonian in matter improves exponentially, with the order of the off-diagonal suppression following the Fibonacci sequence. Second, the claimed 'all matter potentials' reach of that result rests on an unproved cancellation at the solar resonance; the paper is honest about this, but the honest caveat is a real gap.\n\nWhat is new: the Fibonacci recurrence itself, and the explicit strategy 'vacuum (2-3) rotation, then always rotate away the largest off-diagonal element' (LODE) as the best pivot choice. Earlier rotation papers by the same group used fixed sequences or compared with perturbation theory; none identified the Fibonacci rate or systematically compared pivot-selection strategies. The order-counting argument in Sec. 2.2 is correct under the stated O(1) diagonal-gap assumption, and the figures confirm the predicted hierarchy across a wide range of Ye-rho-E. There are no fitted parameters; the rotation angles are algebraic functions of the Hamiltonian and the numerical inputs are standard global-fit parameters. The citation pattern to their own earlier work is appropriate—it describes the method, not pre-empts the new result.\n\nSoft spots. The global optimality of LODE is asserted from numerical exploration, not proved; one can imagine a counterexample elsewhere in parameter space. More importantly, the solar-resonance case breaks the Sec. 2.2 order-counting because the diagonal gap is O(epsilon), and the rescue relies on a numerator-denominator cancellation in the eigenvector-correction metric (Eq. 22). The paper states that this occurs only for a special sequence, and demonstrates it numerically, but does not turn it into a proof that the metric stays at order epsilon^{F_N} for all subsequent rotations and all parameter values. At exactly the solar resonance the denominator in Eq. 27 vanishes, and the degenerate-subspace handling before the next rotation is not specified. These are genuine soft spots, but they are localized: away from that resonance the Fibonacci claim rests on solid ground, and the authors flagged the caveat themselves rather than concealed it. No code is provided, so the numerical claims are not independently reproducible without effort.\n\nWho this is for: neutrino phenomenologists who want fast, compact high-precision diagonalization, and method people interested in structured Jacobi rotations. The paper is worth a serious referee. My recommendation to the editor: send it out; ask the referee to focus on the solar-resonance cancellation and to request that the authors provide code or a more formal argument for that case. The central claim is likely right, but the paper would be stronger if the resonance caveat were either proven or stated as a limit of the method.","headline":"Useful, honest paper: the Fibonacci convergence of the rotation method is real and well demonstrated, but the solar-resonance rescue is numerically shown, not proven.","tokens_in":12280,"tokens_out":1719,"would_cite":true,"duration_ms":17331,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["14.60.Pq"],"model":"deepseek-v4-flash","headline":"A simple rotation sequence diagonalizes neutrino oscillations in matter with errors that shrink at a Fibonacci rate, beating perturbation theory.","keywords":["neutrino oscillations","matter effects","Hamiltonian diagonalization","rotation method","Jacobi method","Fibonacci convergence","solar resonance","perturbation theory"],"falsifier":"At a matter potential just above the solar resonance, for example Ye rho E = 0.250 GeV g/cm3, implement the LODE rotation sequence listed in the paper and compute max_{j>k} |2E (H1)_jk / $\\Delta$ lambda_jk| after each rotation; the paper predicts a suppression to about $10^{-21}$ after seven rotations, so a failure to reach comparably fast suppression would falsify the Fibonacci convergence claim.","tokens_in":11212,"feed_emoji":"🌀","tokens_out":5040,"duration_ms":54755,"temperature":0.7,"pith_summary":"The paper claims that repeated 2x2 rotations of the three-flavor neutrino Hamiltonian in matter converge exponentially fast, with the error order following the Fibonacci sequence. This matters because the exact diagonalization of the matter Hamiltonian is analytically unwieldy and ordinary perturbative expansions degrade near level crossings. The authors identify a concrete optimal recipe: first apply the vacuum (2-3) rotation, then at each step rotate away the largest remaining off-diagonal element. With this recipe, a few rotations suppress the off-diagonal Hamiltonian to order epsilon to a Fibonacci power, yielding high-precision oscillation probabilities at constant per-step cost.","feed_headline":"Fibonacci-sequence rotations crack neutrino oscillations in matter","feed_subtitle":"After one vacuum rotation, picking the largest off-diagonal entry each step beats perturbation theory.","key_machinery":"The load-bearing mechanism is the Fibonacci recursion of rotation orders. At each step one selects a 2x2 submatrix of the Hamiltonian, diagonalizes it with a unitary rotation, and repeats; the pivot is chosen as the largest off-diagonal element (the LODE strategy). The recursion works because a rotation with sine of order epsilon^a converts the existing off-diagonal scales epsilon^a and epsilon^b into new scales epsilon^b and $epsilon^{{a+b}}$, so the leading order after consecutive rotations follows Fibonacci numbers. Near the solar resonance, the argument is saved by a cancellation between the numerator and denominator of the correction metric, which the paper emphasizes occurs only for the special sequence of rotations it identifies.","core_discovery":"The central discovery is that, after an initial vacuum (2-3) rotation, the off-diagonal entries of the neutrino Hamiltonian are hierarchically ordered, and a pivot rule that always selects the largest off-diagonal element produces a Fibonacci recursion in the orders of smallness. If the leading off-diagonal scales are epsilon^a and epsilon^b, a rotation with angle of order epsilon^a creates new off-diagonal terms of order epsilon^b and $epsilon^{{a+b}}$; hence after each rotation the leading order is the sum of the two previous orders, which is the Fibonacci rule. For the fully populated case with a=b=1, the off-diagonal Hamiltonian after N rotations scales as $epsilon^{{F_{N+1}}$}, the eigenvector corrections scale as $epsilon^{{F_{N+1}}$}, and the eigenvalue corrections scale as $epsilon^{{2F_{N+1}}$}. The paper also finds that near the solar resonance, where the diagonal gap is small, a specific rotation sequence produces a numerator-denominator cancellation in the correction metric, and numerical checks confirm that the largest-off-diagonal strategy remains superior to the largest-rotation-angle strategy for all matter potentials.","pith_inferences":["The Fibonacci counting argument is not specific to neutrinos: any nearly diagonal Hermitian matrix with hierarchical off-diagonal entries and order-one diagonal gaps should exhibit a similar rotation-sequence convergence, which may be useful in other coupled-oscillator or quantum-mechanical contexts.","The special cancellation at the solar resonance suggests that a fully analytic proof of Fibonacci convergence at all matter potentials may exist, but the paper only verifies the cancellation numerically, so establishing such a proof would strengthen the general claim.","The sharp contrast between the largest-off-diagonal and largest-rotation-angle pivot rules hints that pivot choice in iterative diagonalization can matter much more than typical matrix-analysis folklore suggests, which could inform numerical algorithms outside neutrino physics."],"forward_implications":["After a handful of rotations, the off-diagonal Hamiltonian is suppressed to order epsilon^{F_{N+1}}, so the eigenvector errors reach precision such as 10^-8 with only four matter rotations at typical energies near the solar resonance.","Because each additional rotation is a single 2x2 diagonalization with constant complexity, the rotation method becomes increasingly more efficient than perturbation theory as the desired precision grows.","The optimal strategy begins with the vacuum (2-3) rotation, which makes the Hamiltonian real and leaves the matter potential unchanged; the largest-off-diagonal pivot rule then performs at least as well as the largest-rotation-angle rule for all energies.","The Fibonacci convergence rate is expected to extend beyond three-flavor oscillations, including scenarios with non-standard interactions or additional sterile neutrinos."],"supporting_citations":[{"why":"Introduces the matter potential that makes the neutrino Hamiltonian non-trivially diagonalizable.","marker":"[1]"},{"why":"The original Jacobi rotation method that this paper's sequential-rotation scheme is built on.","marker":"[3]"},{"why":"A prior analytical approximation applying rotation methods to neutrino matter effects at large theta_13.","marker":"[8]"},{"why":"Provides the compact perturbative expressions that serve as the baseline for comparing rotation convergence with perturbation theory.","marker":"[11]"},{"why":"Earlier work by the authors comparing rotations with perturbative expansions, which the present paper extends by identifying Fibonacci convergence.","marker":"[13]"},{"why":"Supplies the argument that eigenvalue errors are twice the order of the off-diagonal Hamiltonian errors.","marker":"[20]"},{"why":"Defines Delta m^2_ee, the mass-squared scale used throughout the neutrino-sector analysis.","marker":"[24]"},{"why":"Provides the global-fit neutrino parameters used in the numerical comparisons.","marker":"[25]"}],"fun_headline_variants":["Neutrino oscillations solved via Fibonacci-fast rotations","Fibonacci convergence: new pivot rule for neutrino matter","Largest off-diagonal pivot yields Fibonacci speed for neutrinos","Fibonacci fast scheme beats perturbation theory for neutrinos","Fibonacci pivot rule cracks neutrino matter oscillations"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The Fibonacci rate assumes that the two diagonal entries selected for each rotation differ by an order-one amount in units of the small parameter, so the rotation angle inherits the size of the off-diagonal element it removes; at the solar resonance this gap is itself small, and the claimed rate survives only through a numerator-denominator cancellation that is verified numerically, not derived.","fun_headline_variants_meta":{"raw":{"variants":["Neutrino oscillations solved via Fibonacci-fast rotations","Fibonacci convergence: new pivot rule for neutrino matter","Largest off-diagonal pivot yields Fibonacci speed for neutrinos","Fibonacci fast scheme beats perturbation theory for neutrinos","Fibonacci pivot rule cracks neutrino matter oscillations"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000565,"raw_usage":{"total_tokens":2654,"prompt_tokens":898,"completion_tokens":1756,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":514,"completion_tokens_details":{"reasoning_tokens":1684}},"tokens_in":514,"tokens_out":1756,"duration_ms":14171,"temperature":1.0,"reasoning_tokens":1684,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T05:03:41.678467+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"At a matter potential just above the solar resonance, for example Ye rho E = 0.250 GeV g/cm3, implement the LODE rotation sequence listed in the paper and compute max_{j>k} |2E (H1)_jk / $\\Delta$ lambda_jk| after each rotation; the paper predicts a suppression to about $10^{-21}$ after seven rotations, so a failure to reach comparably fast suppression would falsify the Fibonacci convergence claim.","supporting_citations":[],"review_version":1}