{"id":"f4fc4857-c9fa-4ce8-8106-1885d7999afd","arxiv_id":"2607.22464","paper_version":1,"verdict":"ACCEPT","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"For two non-identical coupled nonlinear oscillators with power-law decaying perturbations, resonant phase-locked solutions are stable exactly when the averaged pendulum-like system has a stable equilibrium (π<0, λ'ν<0, ϑ<0).","lead":"This paper proves conditions under which two coupled nonlinear oscillators, with coupling and perturbations decaying over time, lock into a resonant synchronized state with energies near fixed levels. The result is a rigorous mathematical criterion for when such phase-locked regimes form and when they do not.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Theorem 2's π(ψ0)>0 instability condition fails at α=1 for 0<π<1/2, since E1=t^{-1}ρ² decays even when ρ grows.","rationale":"The reader identified the equilibrium-preserving assumption as the weakest point, but that limitation is explicitly acknowledged and is a scope restriction rather than a flaw in the argument. The load-bearing concern is a concrete mathematical overclaim in Theorem 2: the instability condition π(ψ0)>0 ignores the E1=t^{-α}ρ² rescaling. For α=1, amplitude growth ρ∼t^{π} translates to energy decay E1∼t^{2π-1} when π<1/2, and the provided examples can be tuned so that phase and E2 remain bounded. This is a falsifiable counterexample to the stated theorem, not merely a missing proof detail. The same boundary value α=1 also exposes an internal inconsistency: Lemma 2's construction refers to averaged coefficients Π3–Π5, but Theorem 1 permits only N≤2 for α=1. The positive stability result may survive with a corrected threshold and a repaired α=1 proof, so the appropriate disposition is conditional acceptance rather than outright rejection. The proposed numerical test would settle whether the π>0 instability claim is indeed false as stated.","tokens_in":28462,"tokens_out":42026,"duration_ms":446089,"concrete_test":"Numerically integrate system (1) with (10), α=1, ω1=1/2+E, ω2=1−E, parameters a11=0.4, a12=0, a21=0, a22=1, b12=−1, b21=1 (other b=0), initial E1=10^{-6}, E2−A=0, φ1−φ2=π/2 at t_s=10. If E1 stays below, say, 10^{-4} and the phase difference remains bounded while ρ grows, Theorem 2's first instability condition is refuted. Also rederive Lemma 2 for α=1 using only the N=2 model system permitted by Theorem 1; if Π3–Π5 are required, the stability proof is incomplete as written.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The weakest point is the instability half of Theorem 2, not the f1=O(E1) assumption. Bullet (i) says π(ψ0)>0 forces escape from any small neighborhood of the resonant state. But in the averaged variables E1=t^{-α}ρ² (Eq. 11). For α=1, the truncated equation dρ/dt=t^{-1}πρ gives ρ∼t^{π}, so E1∼t^{2π-1}. For 0<π<1/2 this decays to 0. The proof in §6 only shows the auxiliary deviation y1 grows; it does not imply the physical combination E1 grows. In the model example (10) with a11=0.4, a21=0, a22=1 and b1,2=-b2,1, one has π(ψ0)=0.1>0, λ'(ψ0)ν<0, ν2,0(ψ0)=0 and λ2,1=0; the (v,ψ) subsystem is a bounded oscillator, so the original variables stay within O(δ). This contradicts bullet (i). A related α=1 gap: Lemma 2's stability construction uses Π3–Π5, although Theorem 1 restricts N≤2α^{-1}=2 for α=1.","agreement_with_reader":"disagree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies a system of two coupled, non-identical, non-isochronous oscillators with perturbations and coupling that decay as power laws in time. After an averaging transformation, the author derives a model system in new variables E1=t^{-alpha}R^2, E2=A+t^{-alpha/2}v, and a slow phase. The main results, Theorems 1–3, give conditions on averaged coefficients under which resonant phase-locked solutions with E1 near 0 and E2 near a resonant level A are stable or unstable. The proofs combine an averaging transformation with Lyapunov-function analysis, and the theory is illustrated on two Duffing-type examples.","tokens_in":28779,"tokens_out":19234,"duration_ms":220832,"significance":"If correct, the paper would provide explicit, parameter-free stability criteria for resonance capture and phase locking in a genuinely nonautonomous, non-small-parameter setting. The formal structure is attractive: Theorem 1 constructs an averaging transformation with stated remainder orders, and the Lyapunov-function approach is standard and potentially transferable. The examples are useful and the numerics appear to illustrate the intended regimes. However, several load-bearing statements in Theorem 2 are not supported by the proofs as written, and at least one claimed instability criterion is contradicted by the model system itself. These issues affect the central advertised conclusions, not just the presentation.","major_comments":[{"comment":"The condition pi(psi0)>0 is not sufficient for escape in the original variables. From (11) and the truncated equation (18), if pi>0 then rho(t)~t^{pi}, so E1=t^{-alpha}rho^2~t^{2pi-alpha}. Thus for 0<pi<alpha/2, E1 tends to 0 even though rho grows. For example, in system (10) with alpha=1/2, a11=0.4, a21=0, a22=1, b1,2=-b2,1, one has pi(psi0)=0.1>0, lambda'(psi0)nu<0, nu_{2,0}(psi0)=0, lambda_{2,1}=0; the (v,psi) subsystem is a bounded oscillator and E1 decays. The proof in §6 only shows that the auxiliary variable y1, a deviation in rho, grows; it does not show that the physical combination t^{-alpha}rho^2 grows. Bullet (i) is therefore false as stated.","section":"§2, Theorem 2 bullet (i); §6"},{"comment":"The asymptotic construction for alpha=1 uses coefficients Pi_3, Pi_4 and Pi_5, e.g. X1=Pi_3(0,0,psi0), X2=Pi_4(0,0,psi0), X3=Pi_5(0,0,psi0) in (38). However Theorem 1 restricts 1<N<=2alpha^{-1}=2 when alpha=1, so the model system (17) contains only the terms through K=2: Pi_2, Lambda_1, Lambda_2, Omega_1, Omega_2. The higher-order coefficients Pi_3--Pi_5 are not defined for alpha=1 by the stated theorem. The alpha=1 stability construction, and consequently the alpha=1 part of Theorem 2's first bullet, is unsupported unless the theorem is extended or these coefficients are defined separately.","section":"§5, Lemma 2, alpha=1 case"},{"comment":"The proof of Theorem 2 controls |y(t)| in the averaged variables rho,v,psi starting from a small ball |y(t_s)|<=delta. But the theorem's hypotheses and conclusions are stated in terms of E1,E2,phi. Since E1=t^{-alpha}rho^2, the condition |E1(t_s)|<=delta only implies |rho(t_s)|<=t_s^{alpha/2}sqrt(delta), which is not small for large t_s. The Lyapunov argument in §6 therefore does not cover the stated basin, and no uniform-in-t_s stability statement follows. For alpha=1 and -1/2<pi(psi0)<0, a solution with E1(t_s)=delta, E2(t_s)=A, theta(t_s)=psi0 has E1(t)≈delta(t/t_s)^{2pi-1}; the resulting phase drift is approximately eta1 delta t_s, which is unbounded as t_s grows. This is incompatible with the ε-δ formulation of Theorem 2's first bullet. The theorem needs either a reformulation in averaged variables or a genuinely different argument in the original variables.","section":"§2 and §6, original-variable vs averaged-variable basin"}],"minor_comments":[{"comment":"Several crucial steps are delegated to 'it can easily be checked', especially the identities (24) and the remainder estimates (16). Since these support the averaging transformation, please expand these computations or provide a supplementary file with the details.","section":"§3, proof of Theorem 1"},{"comment":"The inequality 'a1,1 < a2,1 a2,1/a2,2' appears to be a misprint. From pi(psi0^+)<0 the condition should presumably involve a1,2 a2,1/a2,2 rather than a2,1^2/a2,2. Please check and correct.","section":"§7.1, Example 1"},{"comment":"In the definition of the Lyapunov function L(z), the integration variable is written as ψ and the lower limit as ψ0; this should be a dummy variable z3 to avoid confusion with the state variable.","section":"§4, proof of Lemma 1"}],"recommendation":"reject","confidential_remarks":"The framework is promising and the averaging construction is nontrivial, but Theorem 2, the central result, is false as stated for parts of the parameter range, and the proof does not connect the averaged-variable basin to the original-variable hypothesis. A substantial reworking of the theorem statements and proofs would be needed before the paper could be considered further."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper gives the first rigorous stability conditions for resonant phase-locking in two coupled non-identical oscillators with power-law decaying perturbations. That is a real advance over the planar case in [24], and the averaging-plus-Lyapunov structure is coherent. The worked Duffing examples are useful and honestly presented. I think the main stability theorem is likely correct for α<1.\n\nBut there is a load-bearing gap in the α=1 case, and the reader's report misses it. In Theorem 2, bullet (i) says π(ψ0)>0 forces escape from any small neighborhood of the resonant state. The proof shows the auxiliary variable y1 grows, but the physical energy is E1 = t^{-α}ρ². For α=1 and 0<π<1/2, the truncated equation gives ρ∼t^π, so E1∼t^{2π-1} → 0. The same issue appears in the model example (10): with a11=0.4, a21=0, a22=1, b1,2=-b2,1, one has π(ψ0)=0.1>0 while the (v,ψ) subsystem is a bounded oscillator, so the original variables stay within O(δ). This directly contradicts bullet (i). The flaw is confined to α=1 and small positive π, but that is a genuine regime.\n\nThere is a second problem in the same corner. Lemma 2's α=1 stability construction uses Π3–Π5, but Theorem 1 restricts N≤2α^{-1}=2 for α=1, so those terms are not in the model system (17). The asymptotic ansatz is being substituted into the wrong equations. The stability claim may be repairable, but the proof as written does not work.\n\nFor α<1, the instability argument is much more plausible, because exponential growth in ρ overpowers the t^{-α} prefactor. So the paper is not fatally wrong everywhere; it is a good framework with an overreaching theorem.\n\nWho should read it? People working on asymptotic theory of nonautonomous oscillator systems will want to know the averaging construction and the example analysis. It deserves a serious referee, but the referee should require a correction of the α=1 statements, not just cosmetic changes. I would not cite it in its current form until the α=1 issue is fixed.","headline":"A serious extension of the planar theory, but the α=1 instability claim overreaches: ρ-growth does not imply energy growth, and the α=1 stability proof uses terms outside the stated model system.","tokens_in":29209,"tokens_out":6871,"would_cite":false,"duration_ms":73668,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["34C15","34C29","34D20","34E10"],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper claims that decaying perturbations can still create stable phase-locked resonance in two nonidentical oscillators, and it gives explicit sign conditions on averaged perturbation coefficients that decide stability.","keywords":["resonance","phase locking","coupled oscillators","decaying perturbations","averaging method","Lyapunov functions","asymptotically autonomous systems","Duffing oscillators"],"falsifier":"Take the model system of Example 1 with parameters satisfying the three stability signs and measure E1(t), E2(t), and φ1(t)-φ2(t) from small initial data; if the phase difference repeatedly slips by 2π or the energies leave a small tube around (0,A), then Theorem 2 is false. Alternatively, choose parameters with ϑ(ψ0)=0 and check whether the predicted Lyapunov-neutral behavior appears, testing the sharpness of the ϑ<0 requirement.","tokens_in":28367,"feed_emoji":"🔗","tokens_out":3953,"duration_ms":44183,"temperature":0.7,"pith_summary":"The paper studies two coupled nonlinear oscillators whose mutual coupling and external perturbations both decay as a power law t^{-α}. It asks whether such fading influences can still lock the oscillators into resonance, so that the first energy stays near zero, the second near a resonant level A, and the phase difference approaches a constant ψ0. The answer is yes under three explicit sign conditions on averaged perturbation coefficients: π(ψ0)<0, λ'(ψ0)ν<0, and ϑ(ψ0)<0. These conditions arise from an averaged three-dimensional model and are proved using Lyapunov functions; when any one of the signs is reversed, the resonant regime is unstable or absent. A sympathetic reader would care because many real synchronization settings involve coupling or forcing that weakens over time, and this paper turns resonance capture into an explicit, checkable criterion.","feed_headline":"Three sign conditions decide when decaying coupling locks phases","feed_subtitle":"For two nonidentical oscillators with t^{-α} forcing, stable resonance is predicted exactly when these averaged coefficients align.","key_machinery":"The key machinery is the averaging transformation (11)-(12) with the near-identity change of variables (22)-(25): it removes the fast phase φ to any finite order and produces the truncated system (17) with coefficients Π_K, Λ_K, and Ω_K. In the leading terms, Π_2=ρπ(ψ), Λ_1=λ(ψ), and Ω_1=νv; the sign of π(ψ0) controls the radial direction, the pair (λ'(ψ0),ν) controls a pendulum-like linearization, and ϑ(ψ0)=λ_{2,2}(ψ0)+ν'_{2,0}(ψ0) controls the next-order correction. Stability is established by explicit Lyapunov functions for the stable case and Chetaev-type functions for the unstable cases.","core_discovery":"The central claim is that the survival of a resonant phase-locking regime is controlled by three averaged quantities. After an averaging change of variables, the leading dynamics of the two oscillator energies and the phase difference reduce to a three-dimensional truncated system whose zero equilibrium (ρ=0, v=0, ψ=ψ0) is stable exactly when π(ψ0)<0 and λ'(ψ0)ν<0, with λ'(ψ0)≠0 guaranteed by a transversality assumption. If, in addition, ϑ(ψ0)<0, this equilibrium generates an asymptotically stable solution of the model system and, by persistence, a stable solution of the original four-dimensional system with E1(t)→0, E2(t)→A, and φ1(t)-κφ2(t)/κ→ψ0. If any of the three signs is reversed, or i","pith_inferences":["The author leaves open the case where the first oscillator is directly excited (f1=O(1) rather than O(E1)); this is a natural test of the boundary of the theory.","The same three-sign structure may generalize to resonances of higher order or to chains of oscillators, where each pair would contribute its own π, λ, and ϑ.","One could test the sharpness of the ϑ=0 boundary numerically; the proof predicts neutral behavior only in a measure-zero parameter set.","The connection to zero-Hopf bifurcation noted in the conclusion suggests that stable resonance capture here is a nonautonomous analogue of a fold-Hopf stability exchange, which the author flags as future work."],"forward_implications":["If Theorem 2 is correct, a phase-locked resonant solution exists and is stable for every sufficiently small initial deviation, for arbitrarily late start times.","The stability check is finite: compute the three averaged coefficients from the perturbation expansions and test their signs.","Systems where λ(ψ)≠0 for all ψ cannot support resonant phase-locking; their phase difference grows without bound (Theorem 3).","Both the stable-locking and phase-drift scenarios appear in the two worked examples, including a system of non-identical Duffing oscillators with decaying coupling.","The averaged model is equivalent to a pendulum with decaying forcing, so classical pendulum intuition transfers to the coupled-oscillator setting."],"fun_headline_variants":["Three signs rule phase-locking under decaying coupling","Stable phase-locking needs three sign alignments","Decaying coupling: three conditions for phase-lock","Phase-locking survives if three signs agree","Three sign checks predict resonance lock stability"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The perturbation must vanish with the first oscillator's energy, meaning f1=O(E1) and f_{1,k,0}=f_{1,k,1}=0; if the first oscillator is directly forced, the averaged model changes structure and the phase-locking conditions no longer apply.","fun_headline_variants_meta":{"raw":{"variants":["Three signs rule phase-locking under decaying coupling","Stable phase-locking needs three sign alignments","Decaying coupling: three conditions for phase-lock","Phase-locking survives if three signs agree","Three sign checks predict resonance lock stability"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000146,"raw_usage":{"total_tokens":993,"prompt_tokens":690,"completion_tokens":303,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":434,"completion_tokens_details":{"reasoning_tokens":233}},"tokens_in":434,"tokens_out":303,"duration_ms":3689,"temperature":1.0,"reasoning_tokens":233,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-01T04:38:27.431180+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take the model system of Example 1 with parameters satisfying the three stability signs and measure E1(t), E2(t), and φ1(t)-φ2(t) from small initial data; if the phase difference repeatedly slips by 2π or the energies leave a small tube around (0,A), then Theorem 2 is false. Alternatively, choose parameters with ϑ(ψ0)=0 and check whether the predicted Lyapunov-neutral behavior appears, testing the sharpness of the ϑ<0 requirement.","supporting_citations":[],"review_version":1}