{"id":"44903d06-7c9d-47bb-8ce6-f2a22f4d6afa","arxiv_id":"2501.16774","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":4.0,"correctness_risk":"high","formal_verification":"none","parameter_count":2,"one_line_summary":"Time-dependent quantum Monte Carlo with B-splines and spherical coordinates yields helium ground-state energies near exact values and predicts enhanced ionization from electron correlation.","lead":"This paper applies the time-dependent quantum Monte Carlo method to three-dimensional helium atoms in laser fields, solving thousands of coupled single-electron Schrödinger equations alongside Monte Carlo walkers. It reports ground-state energies and ionization dynamics for para- and ortho-helium, claiming correlated electrons enhance ionization.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The central claim rests on an unproven mixed-state ansatz: Eq. (3) per-replica product wavefunctions plus KDE averaging cannot reproduce pure-state correlation, and the below-exact energies indicate bias.","rationale":"The reader identified the same load-bearing assumption: the TDQMC equations are asserted, not derived. I agree. I tried to find a more specific internal inconsistency and found the mixed-state/product-ansatz issue: Eq. (3) restricts each replica to a single Slater determinant, and the KDE average over replicas is a mixture. This is not just a missing derivation; it is a structural reason why the method cannot, in general, reproduce pure-state correlation and coherence. The below-exact ground-state energies strengthen this: a method that claims to solve the TDSE should not systematically underestimate the energy. I am not accusing the author of anything; the method may be a useful stochastic mean-field tool, but the paper's central claim 'correlated electron dynamics' and 'fully accounted for' overstates what is shown. The paper does have strengths: the B-spline basis and the TDHF agreement for the uncorrelated limit are reasonable sanity checks, and the scaling argument is plausible. However, those do not validate the correlated case. The proposed 1D benchmark is the minimal check that would settle the concern. Hence I keep the reader's CONDITIONAL verdict: the work is potentially interesting but needs a derivation or benchmark before the central claims can be accepted.","tokens_in":10802,"tokens_out":6731,"duration_ms":65414,"concrete_test":"Apply the identical TDQMC algorithm to a one-dimensional two-electron atom (soft-Coulomb helium) for which an exact grid TDSE solution is cheap, and compare: (i) ground-state energy as M→∞ and σ→0; (ii) survival probability and reduced one-body density under a few-cycle laser pulse. Require the TDQMC ground-state energy to approach the exact value from above (or within statistical error) as M grows, and the ionization curve to match the exact TDSE within a stated tolerance (e.g., 5%). If the energy converges below exact or the ionization differs by more than the TDHF versus exact difference, the ansatz is not equivalent to the true correlated dynamics. This directly tests whether the KDE effective potential and the product ansatz can reproduce genuine two-electron correlation.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Equations (3)-(7) implement a mixed-state mean-field ansatz, not a solution of the many-body TDSE (1). For each replica k, Ψ_k in Eq. (3) is an antisymmetrized product of single-particle guide waves; the many-body density (Eq. 26) is then a convex average over these product densities. This is a statistical mixture of rank-one product states with no inter-replica phase coherence. Since the true helium ground state is a pure, strongly correlated two-electron state, its off-diagonal density-matrix elements and time-dependent coherences are not representable by any such mixture. Moreover, Eq. (5) gives each guide wave a nonlinear, state-dependent effective potential; the ensemble of such equations does not reduce to the linear unitary Eq. (1). The only quantitative evidence for correlation is the KDE-smoothed potential (Eq. 6) with bandwidth from Silverman's rule (Eq. 25), a statistical heuristic whose dynamical relevance is asserted, not derived. The computed para-helium energy (-2.91 a.u.) lies below the exact nonrelativistic value (-2.9037 a.u.), which is impossible for a rigorous variational or exact method and indicates the energy estimator (Eq. 28) is biased by the KDE. Thus the central claim of 'correlated electron dynamics' is not supported by a derivation or an exact benchmark.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript applies the time-dependent quantum Monte Carlo (TDQMC) method to three-dimensional para- and ortho-helium. The method represents the many-body state by per-walker guide waves obeying coupled single-particle time-dependent Schrödinger equations (Eq. (5)), with electron-electron interactions smoothed by kernel density estimation (Eqs. (6)-(7), (25)). The paper reports ground-state energies -2.91 a.u. and -2.2 a.u. for para- and ortho-helium, respectively, compared with -2.86 and -2.17 a.u. for Hartree-Fock, and presents time-dependent ionization and dipole-moment curves that show enhanced ionization in the correlated TDQMC calculations. A spherical-coordinate B-spline implementation is described.","tokens_in":11081,"tokens_out":4673,"duration_ms":42341,"significance":"If the TDQMC representation were rigorously grounded, the method would offer a polynomial-scaling classical algorithm for correlated multi-electron dynamics, which is an important open problem. The manuscript's concrete algorithmic contribution—solving up to 20,000 coupled 3D TDSEs with B-splines and reporting specific energies and ionization dynamics—is nontrivial. However, the central claims rest on an unproven ansatz: the representation of the many-body state as an antisymmetrized product of per-walker guide waves with KDE-smoothed effective potentials is asserted, not derived from the many-body TDSE. The heuristic bandwidth in Eq. (25) and the below-exact ground-state energies further weaken the evidence. The paper would be significant if the method were validated, but the current evidence is insufficient to establish the central claim.","major_comments":[{"comment":"The central representation is asserted rather than derived. Eq. (3) expresses each replica of the state as an antisymmetrized product of single-particle guide waves, and Eq. (26) defines the many-body density as a kernel-smoothed average over walker positions. This is a statistical mixture of product states; it cannot represent the off-diagonal coherences or the full two-body correlation of the pure state that solves Eq. (1). Equations (5)-(7) are nonlinear, state-dependent coupled equations, so the ensemble does not reduce to the linear unitary TDSE. The paper needs either a derivation of this ansatz from Eq. (1) with stated approximations, or a benchmark against an exact solution of a simple two-electron model where the full wavefunction is known, before the ground-state and ionization results can be interpreted as solutions of the many-body TDSE.","section":"Sec. 2, Eqs. (3)-(7) and Eq. (26)"},{"comment":"The reported para-helium ground-state energy of -2.91 a.u. is below the exact nonrelativistic energy of -2.9037 a.u. by 0.006 a.u.; the ortho-helium value of -2.2 a.u. is also below the exact value of approximately -2.175 a.u. For an exact or variational method this is impossible, and it indicates that the energy estimator in Eq. (28) is biased by the KDE bandwidth used in Eq. (25). Although the reported run-to-run fluctuation of 0.05 a.u. makes the deviation comparable to the noise, the point estimate is still on the wrong side of the variational bound. Please quantify the bias as a function of the walker number M and the bandwidth sigma, and show that the estimator converges to the exact energy from above in the joint limit of large M and vanishing sigma.","section":"Sec. 4, ground-state energies"},{"comment":"The nonlocal correlation length sigma_j^k(r,t) is obtained from Silverman's rule (Ref. 19), a statistical density-estimation heuristic. This parameter controls the effective electron-electron potential in Eqs. (6)-(7) and the energy estimator in Eq. (28). The paper does not demonstrate that the physical results are insensitive to this choice or that Silverman's rule follows from the TDSE. Please provide a convergence study in which sigma is varied by factors of two around the value from Eq. (25), with all other parameters fixed, and show that the ground-state energies and ionization probabilities are stable. Without such a test, the agreement with exact energies could be the result of tuning the smoothing width.","section":"Eq. (25)"},{"comment":"The claim that correlated electrons show enhanced ionization due to electron-electron repulsion is not yet substantiated. The comparison is between correlated TDQMC, uncorrelated TDQMC (sigma -> infinity), and TDHF. These calculations differ in more than the inclusion of correlation: the effective potential changes form and the KDE smoothing changes. In particular, M1=30 walkers are used in Eq. (6), so the 10% difference in survival probability could be a finite-sampling artifact. Please provide a convergence study with respect to M1 and a controlled comparison in which only the electron-electron interaction is switched off while all other numerical parameters are kept fixed.","section":"Sec. 4, Figs. 2-3"}],"minor_comments":[{"comment":"The color assignments are inconsistent: the caption says blue lines are time-dependent Hartree-Fock and green is conventional Hartree-Fock, while the text says 'green and red lines show the results from the conventional time-dependent Hartree-Fock (TDHF) and from the uncorrelated TDQMC.' Please clarify which curve corresponds to which method.","section":"Sec. 4, Fig. 2 caption and text"},{"comment":"Eq. (12) writes the potential matrix element as an angular integral, while Eq. (13) gives a multipole expansion; the relationship between the two is not explained. Please indicate whether Eq. (13) is used directly in Eq. (12) or serves only as motivation, and define the angular variables consistently with the rotating-frame procedure.","section":"Eqs. (12)-(13)"},{"comment":"The dipole matrix elements D_{l,l'}^m are written for l'=l+1 and l'=l-1, but the signs and normalization are not fully specified for all m values. Please provide the complete expression or a reference that states it, so the reader can reproduce the calculation.","section":"Eqs. (16)-(17)"},{"comment":"There are several typographical errors, including 'consideres' on page 4, 'Hartle-Fock' in the Figure 2 caption, and inconsistent use of 'time-dependent Hartree-Fock' versus 'conventional Hartree-Fock' in the captions. Please proofread the text.","section":"Introduction and figure captions"},{"comment":"The orthonormality condition in Eq. (8) is written for guide waves belonging to different electrons i and j within the same replica, but these functions depend on different spatial coordinates; the meaning of the integral over a single variable r is unclear. Please clarify whether this is an orthogonality condition in the single-particle Hilbert space after tracing out other coordinates, or something else.","section":"Eq. (8)"},{"comment":"The derivation of the screened-potential exchange starting from Hartree-Fock (Ref. 14) is a self-citation and is not described in detail. Please add an external textbook or review reference for the exchange-hole picture and for the KDE bandwidth selection.","section":"References"}],"recommendation":"major_revision","confidential_remarks":"The manuscript's core method is supported almost entirely by self-citations (Refs. 6-8, 11, 14), and no independent benchmark against a known exact correlated wavefunction is provided. The below-exact ground-state energy is a red flag that the KDE-based estimator may be biased. I recommend the editor ask for a rigorous derivation or a falsifiable benchmark, and for a systematic parameter study, before considering publication. The scope of the journal fits the topic, but the paper is not ready in its present form."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague, the paper applies the author's existing TDQMC method to 3D helium in spherical coordinates with B-splines. The genuinely new pieces are the spherical-coordinate implementation, the para/ortho ground-state energies, and time-dependent ionization curves showing a difference between correlated and uncorrelated electrons. The method itself is from the author's earlier work, so this is a technical extension, not a new framework.\n\nWhat it does well: the numerics are serious. Solving 20,000 coupled 3D TDSEs with B-splines, absorbing boundaries, and a split-step scheme is a real computational effort. The ground-state energies are close to exact values, the comparison with TDHF provides a useful baseline, and the paper honestly reports fluctuations and averages over runs.\n\nThe soft spots are not minor. The core equations—the product ansatz for replicas (Eq. 3) and the KDE-smoothed effective potentials (Eqs. 5–6)—are asserted, not derived from the many-body TDSE. The many-body density (Eq. 26) is a product-kernel estimator over walkers, which makes the state a statistical mixture, not a pure correlated state. There is no inter-replica phase coherence, so it is unclear how the method can represent the off-diagonal coherence of the true two-electron state. The KDE bandwidth from Silverman's rule (Eq. 25) is a heuristic smoothing parameter that controls the effective electron-electron interaction; nothing derives its dynamical relevance. The para-helium energy of -2.91 a.u. sits below the exact nonrelativistic -2.9037 a.u., which is impossible for a rigorous variational or exact calculation and points to bias in the energy estimator (Eq. 28). The ionization results are compared to TDHF only, not to an exact TDSE benchmark, so the enhanced-ionization claim is not verified.\n\nI would not dismiss the work. It is a coherent, plausible computational method with real numerical content, but the paper overreaches in calling it correlated electron dynamics without a derivation or a clean exact benchmark. The right path would be: keep the technical contribution, add a connection to the TDSE or demonstrate a case where the method matches a known exact time-dependent result, and either fix or explain the below-exact energies. I would send it to a referee—there is something here worth examining—but I would expect a major revision. A reader interested in new numerical methods for two-electron dynamics might get value from it; a reader looking for rigorous validation will be disappointed.","headline":"Applies the author's TDQMC method to 3D helium with serious numerics, but the correlated-dynamics claim rests on an unproven mixture ansatz and energies that come in slightly below exact.","tokens_in":11614,"tokens_out":3250,"would_cite":false,"duration_ms":29966,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["31.25.-v","02.70.Ss"],"model":"deepseek-v4-flash","headline":"This paper claims that time-dependent quantum Monte Carlo reproduces correlated electron dynamics in three-dimensional helium, including ground-state energies close to exact values and ionization enhanced by electron-electron repulsion.","keywords":["time-dependent quantum Monte Carlo","helium","electron correlation","strong-field ionization","kernel density estimation","guide waves","B-splines","time-dependent Hartree-Fock"],"falsifier":"Run a converged full two-electron time-dependent Schrodinger calculation for helium under the same two-cycle pulse with $E_0=0.4$ a.u. and $\\omega=0.153$ a.u., and compare the survival probability and dipole moment with the correlated TDQMC curves; if the exact correlated survival probability is not roughly 10% below the time-dependent Hartree-Fock result, then the claimed correlation-enhanced ionization is an artifact of the method.","tokens_in":10516,"feed_emoji":"⚛️","tokens_out":8707,"duration_ms":74110,"temperature":0.7,"pith_summary":"This paper tries to establish that time-dependent quantum Monte Carlo (TDQMC) can solve the correlated two-electron dynamics of a three-dimensional helium atom with resources that scale polynomially instead of exponentially. The method solves up to 20,000 coupled single-particle Schrodinger equations for guide waves while Monte Carlo walkers move by first-order guidance equations and supply the electron-electron interaction through kernel-smoothed potentials. If the claim is right, correlated electron dynamics in strong laser fields, which normally requires an exponentially large configuration-space grid, becomes accessible to classical computers for atoms and eventually molecules. The paper's concrete evidence is that the ground-state energies come out close to exact helium values, uncorrelated TDQMC reproduces time-dependent Hartree-Fock, and correlated TDQMC predicts enhanced ionization attributed to electron-electron repulsion.","feed_headline":"Correlated 3D helium dynamics run on 20,000 guide waves","feed_subtitle":"Method lands near exact ground-state energies and links enhanced ionization to electron-electron repulsion.","key_machinery":"The central object is the TDQMC ansatz of Eq. (3): the many-body wavefunction is a replica-indexed antisymmetrized product of single-electron guide waves. Each guide wave obeys its own 3D time-dependent Schrodinger equation, with the electron-electron potential replaced by a Monte Carlo sum over the other electron's walkers, smoothed by a kernel of width $\\sigma_j^{(k)}(t)$ (Eqs. (5)-(6)). The walkers move by the first-order guidance equation (Eq. (4)), and the kernel width is estimated by kernel density estimation over the walker ensemble (Eq. (25)). This smoothing is what converts discrete walker distributions into continuous correlated charge distributions and is described as the nonlocal correlation length that stabilizes the ground state; when $\\sigma$ goes to infinity the effective potential reduces to the Hartree potential. Numerically, spherical coordinates with B-splines in the radial coordinate, coordinate rotation to preserve the azimuthal quantum number $m$, and a split-step solver carry the computation.","core_discovery":"The paper argues that the correlated two-electron dynamics of helium can be captured by representing the many-body wavefunction as an antisymmetrized product of per-walker guide waves and propagating those guide waves with coupled single-body Schrodinger equations, while Monte Carlo walkers guide the wavefunctions and sample densities. In this scheme, electron-electron repulsion enters through a kernel-smoothed effective potential whose bandwidth acts as a nonlocal correlation length. The reported results include a para-helium ground-state energy of $-2.91$ a.u. versus $-2.86$ a.u. for Hartree-Fock, and an ortho-helium ground-state energy of $-2.2$ a.u. versus $-2.17$ a.u., with run-to-run fluctuations near $0.05$ a.u. In an external field with $E_0=0.4$ a.u., correlated TDQMC gives a para-helium survival probability about 10% lower than time-dependent Hartree-Fock, while uncorrelated TDQMC matches TDHF, and the dipole amplitude is largest during the first half-cycle. The author takes this as evidence that tunneling ionization and the correlation-driven enhancement are correctly described.","pith_inferences":["Editorial extension: if the guide-wave ansatz holds, the same reduction could be applied to molecules and to more than two electrons, where full configuration-space grids are intractable; the kernel bandwidth would then become a physical correlation-length parameter rather than only a numerical smoothing scale.","Editorial extension: the reported 0.05 a.u. run-to-run energy fluctuation and the use of only $M_1=30$ walkers in the effective potential suggest a direct convergence test: increasing walker number and varying the kernel bandwidth should leave the ionization curves unchanged if the correlation-enhanced ionization is physical.","Editorial extension: because survival probability is defined by walkers remaining on the numerical grid, boundary placement may influence the quoted 10% enhancement; an exact comparison should use absorbing-boundary-converged ionization yields.","Editorial extension: the method's built-in switch between uncorrelated ($\\sigma\\to\\infty$) and correlated dynamics offers a natural way to isolate correlation effects in other strong-field observables beyond ionization and dipole moments, such as high-harmonic spectra."],"forward_implications":["Correlated TDQMC places the para-helium ground state at $-2.91$ a.u. and ortho-helium at $-2.2$ a.u., a few hundredths of a hartree below the Hartree-Fock values, matching the known pattern that correlation deepens binding.","For para-helium at $E_0=0.4$ a.u., correlated TDQMC predicts about 10% higher ionization than time-dependent Hartree-Fock, while uncorrelated TDQMC tracks TDHF, isolating electron-electron repulsion as the cause of the enhancement.","The solver handles up to 20,000 coupled 3D guide-wave equations with spherical coordinates and B-splines, so the exponential configuration-space scaling of the exact many-body TDSE is replaced by a walker ensemble with polynomial scaling.","In ortho-helium, where the outer 2s electron is weakly bound, correlated TDQMC also shows enhanced ionization at the lower field amplitude $E_0=0.03$ a.u.","Because each guide wave is independent apart from smoothed effective potentials, the TDQMC equations are highly parallel, allowing very large walker ensembles to be propagated in affordable time."],"supporting_citations":[{"why":"Introduces the time-dependent quantum Monte Carlo method with simultaneously evolving walkers and guide waves.","marker":"[6]"},{"why":"Develops the TDQMC formalism applied in the present calculations.","marker":"[7]"},{"why":"Supplies the coupled guide-wave equations and effective electron-electron potentials (Eqs. (5)-(6)) that the paper applies to helium.","marker":"[8]"},{"why":"Provides the multipole expansion of the Coulomb electron-electron interaction used in the potential matrix elements.","marker":"[15]"},{"why":"Provides the B-spline basis set methods used for the radial expansion.","marker":"[17]"},{"why":"Supplies the kernel density estimation bandwidth formula used in Eq. (25) for the nonlocal correlation widths.","marker":"[19]"}],"fun_headline_variants":["20,000 guide waves simulate correlated helium dynamics","Monte Carlo method captures correlated helium ionization","Enhanced ionization from electron repulsion in helium model","Correlated electrons in 3D helium via time-dependent QMC","Helium ionization boosted by correlation in QMC simulation"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the true two-electron correlated state can be replaced by a set of individual wave packets that talk to each other only through smoothed average electron-electron forces, with the smoothing width chosen from the walker data; if that replacement is not equivalent to solving the full two-electron Schrodinger equation, the reported energies and ionization curves do not follow.","fun_headline_variants_meta":{"raw":{"variants":["20,000 guide waves simulate correlated helium dynamics","Monte Carlo method captures correlated helium ionization","Enhanced ionization from electron repulsion in helium model","Correlated electrons in 3D helium via time-dependent QMC","Helium ionization boosted by correlation in QMC simulation"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000773,"raw_usage":{"total_tokens":3411,"prompt_tokens":927,"completion_tokens":2484,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":543,"completion_tokens_details":{"reasoning_tokens":2409}},"tokens_in":543,"tokens_out":2484,"duration_ms":14005,"temperature":1.0,"reasoning_tokens":2409,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T10:48:11.737718+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Run a converged full two-electron time-dependent Schrodinger calculation for helium under the same two-cycle pulse with $E_0=0.4$ a.u. and $\\omega=0.153$ a.u., and compare the survival probability and dipole moment with the correlated TDQMC curves; if the exact correlated survival probability is not roughly 10% below the time-dependent Hartree-Fock result, then the claimed correlation-enhanced ionization is an artifact of the method.","supporting_citations":[],"review_version":1}