{"id":"b3593e04-c85f-4aaf-b03d-c29ea64710b6","arxiv_id":"2608.08108","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"First- and second-order formulas for pole shifts, projector corrections, and transition amplitudes for time-dependent renormalized point interactions in discrete-spectrum systems.","lead":"Quantum mechanics with a point-like force (a delta potential) is usually treated with a fixed force. This paper derives a perturbation theory for forces that change over time, either in strength or position, in systems with a discrete energy spectrum, and gives explicit formulas for energy shifts and transition probabilities with worked examples.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The sphere example's moving-center expansion is not L²-valid: ∂_s ψ_α has coefficients ∼L^{-1/2} and the first-order transition probability diverges as ∑1/L, so the computed amplitudes are meaningless.","rationale":"The reader's weakest assumption was the unproved identification of the frozen spectral operator H(t)=∑E_k(t)P_k(t) with the physically defined time-dependent point-interaction Hamiltonian. That domain-theoretic gap is real, but it is not the most load-bearing problem: the sphere example fails even if the postulate is granted. In that example, the derivative of the instantaneous eigenfunction has expansion coefficients decaying only as L^{-1/2}, so its L² norm diverges logarithmically. Consequently the projector derivative used in the moving-center perturbation theory is not a well-defined operator, and the computed transition amplitudes violate unitarity by giving infinite total first-order probability. This is a concrete, checkable internal inconsistency in a central advertised example. The general formalism may survive under a UV-decay condition (e.g., ∑|⟨φ_m|∂_q ψ_k⟩|²/ω_mk² < ∞), and the 2D oscillator example appears to satisfy such a condition, but the paper neither states nor proves it. Therefore the verdict should remain conditional, with the condition strengthened to require convergence of the moving-center perturbation series; the sphere example must be corrected or removed unless the divergence is resolved.","tokens_in":33500,"tokens_out":26448,"duration_ms":284022,"concrete_test":"For §5.2, set R=M=ℏ=1, initial zonal state ψ_α, and generic off-resonant Ω (e.g., Ω=0.37) at t=1. Compute S_N(t)=∑_{L=1}^{N}∑_{M=±1}|c^{(1)}_{LM}(t)|² using Eqs. (5.81)–(5.82). If S_N grows like C log N for N up to 10^4, the first-order state is not normalizable and the example's transition amplitudes are invalid. Independently, sum |B_L|²/(E_L−E*_α)²; divergence of this sum proves that ∂_s ψ_α (and hence P_α,E) does not exist in L², contradicting Proposition 4.2's use of it in the sphere case.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"Section 5.2's sphere example is internally inconsistent. Equation (5.59) defines B_L ∼ L^{3/2}, and (5.60) gives ∂_s φ_{L,±1}(N)=±B_L. The derivative of the zonal point-interaction state, Eq. (5.76), has free-basis coefficients ⟨φ_{L,±1}|∂_s ψ_α⟩ = ±B_L/[(E_L−E*_α)√N_α] ∼ L^{-1/2}. Hence ∑_{L,M}|⟨φ_{L,M}|∂_s ψ_α⟩|² ∼ ∑ L^{-1} diverges: ∂_s ψ_α is not an L² vector, so the instantaneous projection P_α(s) is not strongly differentiable at s=0. The paper nevertheless uses this derivative to compute first-order amplitudes (5.79)–(5.82); the resulting |c^{(1)}_{L,±1}(t)|² ∼ L^{-1} gives a logarithmically divergent total transition probability. The same UV divergence enters the second-order sums (5.86) and (5.88) because |φ_{l0}(N)|² ∼ l and B_L²/(E_L−E*_α)² ∼ l³/l⁴ = 1/l. Thus, even accepting the spectral-generator postulate (3.7), the perturbation expansion in the sphere example is not an asymptotic expansion in L²: the first-order correction is non-normalizable. The source is the growth of spherical-harmonic derivatives at the pole, which the paper does not check; this is a concrete failure of the claimed computability of transition amplitudes, not merely a missing domain proof.","agreement_with_reader":"disagree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper develops a perturbative framework for time-dependent renormalized point interactions in systems with purely discrete spectrum. Starting from the static spectral representation H = sum_k E*_k P_k, it derives first- and second-order pole shifts and projection corrections for two types of time dependence: a time-dependent renormalized strength (Section 3) and a moving support point (Section 4). Transition amplitudes are then obtained by inserting the expanded Hamiltonian matrix elements into second-order Dyson-type equations. The formalism is illustrated by a sinusoidally modulated coupling, a two-dimensional harmonic oscillator with a moving or circularly moving delta center, a point interaction moving on a great circle of the two-sphere, and a one-dimensional oscillator with a moving delta, where the result is compared with the naive distributional expansion.","tokens_in":33692,"tokens_out":9210,"duration_ms":97871,"significance":"If valid, the framework would be a genuinely useful complement to the existing literature on time-dependent point interactions: it avoids writing the formal delta potential and expresses all corrections in terms of Green functions, heat kernels, and derivatives of the principal function Phi. The algebraic derivation is transparent, the one-dimensional oscillator example correctly reduces to the standard delta-perturbation result, and the selection rules obtained in the two-dimensional oscillator are explicit and physically suggestive. However, the paper's central computational claim is only conditionally established: the identification of the dynamics with the frozen spectral operator H(t) = sum_k E_k(t)P_k(t) is assumed rather than proved, and the sphere example exhibits a concrete failure of L^2 summability in the first-order transition amplitudes. The sphere failure is not a cosmetic issue; it invalidates the advertised computation of transition probabilities in that example and shows that the general moving-center expansion needs additional hypotheses.","major_comments":[{"comment":"The first-order moving-center expansion in the sphere example is not L^2-valid. From Eqs. (5.59)-(5.60), B_L ~ L^{3/2}, while E_L - E*_alpha ~ L^2, so Eq. (5.77) gives <phi_{L,+/-1}|d_s psi_alpha> = +/- B_L/((E_L-E*_alpha) sqrt(N_alpha)) ~ L^{-1/2}. Hence sum_{L,M} |<phi_{L,M}|d_s psi_alpha>|^2 ~ sum_L 1/L diverges logarithmically: d_s psi_alpha is not an L^2 vector and the projector P_alpha(s) is not strongly differentiable at s=0. Consequently the first-order transition amplitudes (5.81)-(5.82) do not define a normalizable state correction; the total first-order transition probability sum_L |c^{(1)}_{L,+/-1}(t)|^2 diverges. The same ultraviolet divergence affects the second-order sums (5.86) and (5.88). This is a concrete failure of the claimed computability of transition amplitudes in a canonical compact example, not merely a missing domain proof. The authors need either to prove the relevant summability from decay of the eigenfunction derivatives or to replace the example with one where the expansion is an asymptotic expansion in L^2.","section":"Section 5.2"},{"comment":"The paper postulates that the non-autonomous dynamics is generated by the frozen spectral Hamiltonian H(t) = sum_k E_k(t)P_k(t). This premise is load-bearing because every transition amplitude in Sections 3-5 is computed from this equation, yet the paper does not prove that this family equals the established time-dependent point-interaction Hamiltonians of Refs. [2-8], whose operator domains depend on time through the moving center and renormalized coupling. In one dimension the form domain may be fixed, but in two and three dimensions this is not automatic. The authors should either prove equivalence with the known non-autonomous point-interaction evolutions, or explicitly state and justify H(t) = sum_k E_k(t)P_k(t) as a defining model assumption. A concrete diagnostic would be to compare the resulting transition amplitudes with the exact propagators available for moving delta-type interactions in Refs. [9-11] or with the solvable three-dimensional moving-point-interaction problems of Refs. [3,4].","section":"Section 3"},{"comment":"The operator expansions (3.21)-(3.22) and (4.34) are presented as identities involving sums over the full spectral family, but no convergence is shown in any operator topology. The paper does not state hypotheses under which the term-by-term differentiation of the spectral representation is legitimate. The sphere example of Section 5.2 shows that these hypotheses are not automatically satisfied: the failure is not a technicality about unboundedness of individual matrix elements but a divergence of the sum of their squares. The general framework should therefore be formulated with explicit conditions on the decay of G0(x,a|E) and its derivatives, or at least with a statement that the expansions are formal at this stage.","section":"Sections 3-4"}],"minor_comments":[{"comment":"The phrase 'Starting again from (M4)' refers to an equation label that does not appear anywhere in the manuscript; it should be a reference to Eq. (4.32) or Eq. (4.34).","section":"Section 4"},{"comment":"In the first sum of Eq. (5.31) the index l is used both for the initial state and as a summation index in V_{mn}V_{nl}J_{mnl}; the summation should run over n, as in Eqs. (5.18)-(5.19).","section":"Section 5.1"},{"comment":"Equation (2.15) contains a stray argument and inequality: the expression '=-G^0_{k,E}(x)<0' is confusing. The quantity Phi^E_k is a number, namely -d/dE G0(a,a|E) evaluated at E*_k, and the x-dependence should be removed.","section":"Section 2"},{"comment":"The text refers to 'Fig. 1', but the figure is not included in the manuscript; either include it or remove the reference.","section":"Section 5.2"}],"recommendation":"major_revision","confidential_remarks":"The algebraic core of the paper is competently written, and the one-dimensional example is a good consistency check. The sphere example, however, is more than a weak illustration: its first-order amplitudes diverge in L^2, which undermines the paper's stated claim to provide computable transition amplitudes for moving point interactions. The authors should also address the status of Eq. (3.7) as a postulate versus a proved equivalence, since the manuscript's title and abstract present the framework as applying to time-dependent singular quantum systems generally. If the sphere example cannot be repaired, the scope of the paper should be narrowed accordingly."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear colleague:\n\nThis paper is a formal perturbation scheme for time-dependent point interactions, built from the frozen spectral data of the static renormalized Hamiltonian. The algebraic core is clean: pole shifts come from Taylor-expanding the principal function, projection derivatives come from the residue formula, and the one-dimensional moving-delta example reproduces exactly the naive δ′(x) expansion. That 1D match is a real check, and the formulas (3.8)–(3.9) and (4.17)–(4.18) hang together internally. As a heuristic toolkit it has something new: nobody has written time-dependent Dyson series directly in the static point-interaction basis.\n\nThe soft spots are not incidental. Equation (3.7) postulates that the dynamics is generated by H(t)=∑E_k(t)P_k(t); there is no proof that this equals the established time-dependent point-interaction Hamiltonians of Refs. [2–8], whose domains move with the center. In 2D/3D this is not a technicality.\n\nThe stress-test note is right, and the failure is more concrete than a missing domain proof. From (5.59)–(5.60), ∂_sψ_α has free-basis coefficients B_L/(E_L−E*_α) ~ L^{-1/2}. If you square and sum over L,M you get ∑ 1/L, divergent: the derivative of the state is not L², so s ↦ ψ_α(s) fails to be strongly differentiable. The paper uses this derivative to compute the first-order amplitudes (5.79)–(5.82), and the resulting transition probabilities are non-normalizable. The individual matrix elements are finite in a distributional sense, but the series for the first-order wavefunction does not converge in L². That is a genuine failure of the worked example, and the 3D moving-center case will be worse. So I agree with the reader's 'conditional' verdict but would move it closer to 'reject as stated': the advertised sphere calculation does not define a Hilbert-space perturbation theory.\n\nCredit where due: the strength-modulation case (Section 5.1) and the 1D oscillator are solid; the degeneracy treatment is honest, though only by example. The ChatGPT note in the acknowledgments makes independent checking of the examples especially wise.\n\nAudience: people who want a quick calculus for these models, with the warning that it is a formal calculus until the generator identification and L²-regularity are settled. It deserves a serious referee—the 1D check alone is publishable and the questions raised are real—but I would not cite it in the next year and would not put it on anyone's reading list before the divergence problem is fixed or localized to the renormalized examples.\n\nRecommendation: send to peer review with a referee instructed to pin down the domain-theoretic status of Eq. (3.7) and to redo the sphere example with explicit L² convergence checks. As written, the central example fails.","headline":"A clean formal expansion with a good 1D sanity check, but the moving-center examples in 2D/3D are not L²-differentiable—the sphere's first-order transition probabilities diverge, so the framework is heuristic until the generator and domain questions are settled.","tokens_in":34373,"tokens_out":6733,"would_cite":false,"duration_ms":69891,"reading_group":"no","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["81Q15","81Q10"],"pacs":[],"model":"deepseek-v4-flash","headline":"The paper develops first- and second-order perturbation theory for time-dependent point interactions by expanding the renormalized Hamiltonian in the static spectral basis, avoiding the formal delta potential.","keywords":["time-dependent point interactions","renormalized resolvent","heat-kernel regularization","time-dependent perturbation theory","moving delta center","principal function","transition amplitudes","harmonic oscillator"],"falsifier":"For the two-dimensional oscillator with a moving renormalized center, solve the time-dependent Schrödinger equation numerically with a narrow regularized potential, take the width to zero, and compare the transition probability into a first odd-oscillator state with the square of (5.127); any disagreement at order $\\varepsilon^2$ that persists under regularization would show that $H(t)=\\sum_k E_k(t)P_k(t)$ is not the Hamiltonian governing the singular system.","tokens_in":33131,"feed_emoji":"⚛️","tokens_out":12562,"duration_ms":117891,"temperature":0.7,"pith_summary":"This paper tries to establish that a time-dependent point interaction—a singular potential supported at a point whose renormalized strength or position varies with time—can be treated by the same spectral perturbation theory used for ordinary Hamiltonians, provided the singular interaction is encoded through its renormalized spectral data rather than through the formal delta function. If the construction is correct, first- and second-order transition amplitudes for these systems follow directly from the static eigenvalues, eigenfunctions, and the principal function $\\Phi$, with no need to write down the singular potential or its time-dependent domain. The paper derives explicit pole shifts, projection corrections, and transition-amplitude equations for both a time-dependent strength and a moving support point, and illustrates them on a sphere, a two-dimensional oscillator, and a one-dimensional oscillator.","feed_headline":"Moving delta potentials get first- and second-order amplitudes","feed_subtitle":"Static eigenvalues and eigenfunctions give the amplitudes, with no formal delta needed.","key_machinery":"The central object is the renormalized principal function $$\\Phi(E,M)=\\frac{1}{\\alpha_R(M)}+\\int_0^\\infty K_s(a,a)\\bigl($e^{{sM}}$-$e^{{sE}}$\\bigr)\\,ds,$$ the denominator of the renormalized resolvent whose zeros are the shifted bound-state energies; for a moving support, $a$ is replaced by the curve $q(t)$ and the diagonal heat kernel is evaluated at $q(t)$. The spectral representation $H=\\sum_k E_k^\\ast P_k$ of the static renormalized Hamiltonian turns the singular operator into ordinary spectral data, and the paper's main machinery is the Taylor expansion of $\\Phi$ and of $P_k(E_k(t),q(t))$ in the small parameters. This produces the pole shifts (3.8)–(3.9) and (4.17)–(4.18), the projection derivatives (3.19)–(3.20), and the Hamiltonian corrections (3.21)–(3.22) and (4.32)–(4.34) that feed the transition-amplitude equations.","core_discovery":"The central claim is that the non-autonomous Hamiltonian of a renormalized point interaction admits the instantaneous spectral representation $H(t)=\\sum_k E_k(t)P_k(t)$, where the $E_k(t)$ are the zeros of the frozen principal function and the $P_k(t)$ are the corresponding rank-one projections built from the static Green function. The paper shows how to expand $E_k(t)$ and $P_k(t)$ in powers of the small time-dependent parameter $\\eta(t)$ (for a varying renormalized strength) or of the support displacement $s(t)$ (for a moving center), obtaining closed formulas for the first- and second-order pole shifts and for the derivatives of the projections. These ingredients are then inserted into the time-dependent Schrödinger equation in the static eigenbasis, yielding first- and second-order transition amplitudes through the amplitude equations (3.33)–(3.36). The entire calculation runs on Green functions, heat kernels, and spectral data; the formal delta potential is never used.","pith_inferences":["I would test the frozen-spectral identification directly by comparing the paper's second-order amplitudes for the two-dimensional oscillator with a numerical solution of the time-dependent Schrödinger equation for a regularized narrow potential, since (5.117), (5.127), and (5.137) give explicit predictions.","If the frozen spectral family does not coincide with the rigorously defined time-dependent point-interaction Hamiltonian in two or three dimensions, the formulas could still describe a slow-motion effective dynamics, but establishing that would require a separate limit argument.","The single-center construction should extend to several moving centers by replacing the scalar principal function with a finite-dimensional principal matrix; whether the order-by-order pole-shift structure survives with off-diagonal couplings is a natural next calculation."],"forward_implications":["For any unperturbed Hamiltonian with purely discrete spectrum, the first- and second-order response to a time-dependent renormalized strength is fixed by the static principal function, its energy and coupling derivatives at the pole, and the static Green function.","For a moving support, the same response is fixed by the Green function and its first and second derivatives along the curve, including the covariant acceleration, so the formulas separate the motion of the energy levels from the motion of the eigenprojectors.","When the principal function is position-independent, as on the two-dimensional sphere by the addition theorem, the energy levels remain stationary to all orders while the first-order transition amplitudes are nonzero; the dynamics is carried entirely by the moving eigenprojectors.","Sinusoidal driving produces the standard resonance denominators $\\omega_{ml}\\pm\\Omega$ and splits the second-order amplitude into iterated, phase, and direct terms, while a delta-kick produces a frozen transition probability after the kick."],"supporting_citations":[{"why":"Establishes the point-interaction framework and resolvent formulas that the paper treats as its starting point.","marker":"[1]"},{"why":"Defines the rigorous non-autonomous point-interaction dynamics whose relation to the frozen spectral family is the comparison target.","marker":"[2]"},{"why":"Supplies the Laplace–Beltrami spectral expansion and covariant derivative formalism used for the sphere and moving-center Taylor expansions.","marker":"[12]"},{"why":"Provides the renormalized resolvent and principal-function construction in two and three dimensions that the whole perturbation theory is built on.","marker":"[16]"},{"why":"Supplies the heat-kernel estimates that justify convergence of the spectral sums and the statement that eigenfunctions vanishing at the support are unaffected.","marker":"[18]"},{"why":"Gives completeness of the renormalized eigenfunctions, which justifies writing the static Hamiltonian as $\\sum_k E_k^\\ast P_k$.","marker":"[19]"},{"why":"Provides the explicit one-dimensional oscillator-with-delta solution used as a consistency check for the non-renormalized case.","marker":"[22]"},{"why":"Supplies the Mehler heat kernel for the two-dimensional oscillator examples.","marker":"[28]"}],"fun_headline_variants":["Perturbative amplitudes for moving point interactions","Spectral perturbation for time-dependent delta potentials","Moving delta without formal delta: second-order amplitudes","Perturbation for moving delta: spectral shifts to second order","Renormalized point interactions: first and second order shifts"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing assumption is that the true dynamics is generated, at every instant, by the operator obtained from the static eigenvalues and eigenprojections taken at that instant's parameter values, and the paper does not prove that this operator is the same as the one defined by the rigorous time-dependent point-interaction construction in two or three dimensions, where the class of wavefunctions on which the Hamiltonian acts depends on time.","fun_headline_variants_meta":{"raw":{"variants":["Perturbative amplitudes for moving point interactions","Spectral perturbation for time-dependent delta potentials","Moving delta without formal delta: second-order amplitudes","Perturbation for moving delta: spectral shifts to second order","Renormalized point interactions: first and second order shifts"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000852,"raw_usage":{"total_tokens":3682,"prompt_tokens":901,"completion_tokens":2781,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":517,"completion_tokens_details":{"reasoning_tokens":2705}},"tokens_in":517,"tokens_out":2781,"duration_ms":24256,"temperature":1.0,"reasoning_tokens":2705,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T00:27:18.811989+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"For the two-dimensional oscillator with a moving renormalized center, solve the time-dependent Schrödinger equation numerically with a narrow regularized potential, take the width to zero, and compare the transition probability into a first odd-oscillator state with the square of (5.127); any disagreement at order $\\varepsilon^2$ that persists under regularization would show that $H(t)=\\sum_k E_k(t)P_k(t)$ is not the Hamiltonian governing the singular system.","supporting_citations":[{"cited_title":"Albeverio, F","cited_arxiv_id":null,"evidence_quote":"Establishes the point-interaction framework and resolvent formulas that the paper treats as its starting point."},{"cited_title":"A limit evolution problem for time- dependent point interactions,","cited_arxiv_id":null,"evidence_quote":"Defines the rigorous non-autonomous point-interaction dynamics whose relation to the frozen spectral family is the comparison target."},{"cited_title":"Rosenberg,The Laplacian on a Riemannian Manifold, Cambridge University Press, Cambridge, 1998","cited_arxiv_id":null,"evidence_quote":"Supplies the Laplace–Beltrami spectral expansion and covariant derivative formalism used for the sphere and moving-center Taylor expansions."},{"cited_title":"On Schr¨ odinger operators modified by δinteractions,","cited_arxiv_id":null,"evidence_quote":"Supplies the heat-kernel estimates that justify convergence of the spectral sums and the statement that eigenfunctions vanishing at the support are unaffected."},{"cited_title":"Completeness relation in renormalized quantum sys- tems,","cited_arxiv_id":null,"evidence_quote":"Gives completeness of the renormalized eigenfunctions, which justifies writing the static Hamiltonian as $\\sum_k E_k^\\ast P_k$."},{"cited_title":"Solution of the one-dimensional harmonic oscillator with a delta- function potential,","cited_arxiv_id":null,"evidence_quote":"Provides the explicit one-dimensional oscillator-with-delta solution used as a consistency check for the non-renormalized case."},{"cited_title":"Thangavelu,Lectures on Hermite and Laguerre Expansions, Mathematical Notes, Vol","cited_arxiv_id":null,"evidence_quote":"Supplies the Mehler heat kernel for the two-dimensional oscillator examples."}],"review_version":1}