{"id":"e0c7fb58-fd22-45d6-9953-7b0240481a9d","arxiv_id":"2603.12844","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"A frequency-resolved Floquet functional renormalization group for the driven Anderson impurity matches perturbative benchmarks and indicates the Kondo cloud survives local periodic driving.","lead":"This paper builds a frequency-resolved functional renormalization group method that works directly in the periodic steady state of driven interacting quantum systems, and applies it to the driven Anderson impurity. It finds the drive broadens the Kondo resonance but leaves the extended Kondo cloud intact, so the many-body singlet partially survives local periodic driving.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Kondo-cloud-intact claim relies on channel-decomposed FRG that the paper's own Appendix A shows misses the exponential Kondo scale; without an exact steady-state benchmark, the physical headline is unsupported.","rationale":"Read in good faith, the paper develops a frequency-resolved Floquet FRG with a channel-decomposed vertex and applies it to the driven SIAM. The formal derivation and numerical implementation are substantial, and the weak-coupling benchmarks against 2PT and GW support the method claim for small U/Δ. The load-bearing weakness is not the method's internal consistency but the use of that method to make a strong physical claim about a nonperturbative quantity. The Kondo scale is exponential in U/Δ, and the channel-decomposed one-loop FRG cannot reproduce that exponential behavior even in equilibrium (Fig. 6b). The authors are transparent about this: Appendix A says all methods fail to capture the exponential scaling, and Section IVC says the FRG could overestimate the drive-induced suppression. The robust conclusion about the Kondo cloud is thus supported only by an approximation whose known error is largest for the very quantity at issue. Agreement with 2PT and GW does not remove the concern, because those are also perturbative/ladder approximations and not exact benchmarks. The paper itself calls for exact verification with a method like Ref. [113]. I agree with the reader that this is the weakest assumption. I would not reject the paper: the method may be useful, the derivation is careful, the weak-coupling checks are real, and the physical claim is marked as needing exact verification. The correct outcome remains CONDITIONAL rather than ACCEPT.","tokens_in":27539,"tokens_out":5195,"duration_ms":50085,"concrete_test":"Perform a numerically exact Floquet steady-state computation for the driven SIAM at the parameters of Fig. 4/5 (U/Δ=5 and 8, Ω/Δ=5, A/Δ=0.5 and 5.0, T_res=0.05Δ), using an established method such as the semigroup influence-matrix approach (Ref. [113]) or time-dependent NRG. Compute m*=1−∂_ω ReΣ^R(ω)|_{ω=0} and G_lin(0) vs Vg. If exact results reproduce the same A-dependent suppression of m* and the partial pinning conductance plateau without Kondo sidebands, the claim survives; if m* is less suppressed or Floquet sidebands appear on the Kondo peak, the FRG conclusion is an artifact.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central physical claim—'periodic driving ... leaves the many-body Kondo cloud largely intact'—is inferred from the effective mass m* (Fig. 4) and the zero-bias conductance (Fig. 5), both produced by the one-loop channel-decomposed FRG of Eq. (48): Γ ≈ Γ0 + P(ω_P)+C(ω_C)+D(ω_D), with no inter-channel mixing. The paper's own equilibrium benchmark, Fig. 6(b) and Appendix A, states that 'all methods fail to capture the exponential scaling' and, for weak drive, 'the disagreement ... is similar to the disagreement in equilibrium'; the text also concedes 'the FRG could overestimate the suppression' of the Kondo resonance. Since the observable used to establish robustness is exactly the low-energy Kondo scale that this truncation misrepresents, the robustness statement is not established. 2PT and GW agreement cannot decide the issue: all three are uncontrolled ladder-type approximations in the Kondo regime. Additionally, 'Kondo cloud' is a nonlocal correlation, but no direct nonlocal spin-spin correlator is computed; it is inferred from local pinning. The claim is therefore plausible but untested.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper develops a Floquet functional renormalization group (FRG) framework formulated directly in the Floquet steady state. The two-particle vertex is approximated by a channel-decomposed frequency-dependent form (Eq. 48), the flow is regularized by a hybridization-flow cutoff with the Katanin substitution, and the flow equations are solved numerically on a nonuniform frequency grid. The method is benchmarked against second-order perturbation theory (2PT) and self-consistent GW for the periodically driven single-impurity Anderson model, with additional tests of static approximations. The authors find agreement among the methods at weak coupling, observe drive-induced broadening of the Kondo resonance, and interpret the persistence of Kondo pinning and the absence of Floquet sidebands on the Kondo feature as evidence that the nonlocal Kondo cloud survives local periodic driving.","tokens_in":27855,"tokens_out":4808,"duration_ms":50350,"significance":"The methodological development is potentially valuable: a frequency-resolved Floquet FRG that retains some structure of the two-particle vertex is a nontrivial extension of earlier static-vertex approaches, and the derivation from exact FRG identities is clearly presented. The numerical implementation is detailed, includes causality checks, and appears to have been carried out with care. If the physical conclusion were quantitatively established, the result that a local drive broadens but does not destroy nonlocal Kondo correlations would be of interest to the driven-quantum-materials community. However, the central physical claim currently rests on the accuracy of a one-loop channel-decomposed approximation in precisely the regime where the paper's own equilibrium benchmark shows that this approximation fails quantitatively. The benchmark against 2PT and GW does not resolve this issue because all three methods are approximate ladder-type resummations and share uncontrolled behavior in the Kondo regime. The significance of the paper therefore hinges on a robustness claim that is plausible but not yet supported by the presented evidence.","major_comments":[{"comment":"The headline claim that periodic driving 'leaves the many-body Kondo cloud largely intact' is not quantitatively supported by the data shown. The evidence for this claim is the effective mass m* (Fig. 4) and the zero-bias conductance (Fig. 5), both obtained from the one-loop channel-decomposed FRG of Eq. (48) without inter-channel mixing. Appendix A and Fig. 6(b) show that this class of FRG fails to reproduce the exponential equilibrium scaling m* ~ exp(U/Δ); the text itself states that 'all methods fail to capture the exponential scaling' and later concedes that 'the FRG could overestimate the suppression' of the Kondo resonance. Since the observable used to infer robustness is precisely the low-energy Kondo scale that this truncation misrepresents, the robustness statement is not established. A benchmark against a numerically exact steady-state method (e.g., the semigroup influence-mat","section":"§IVC, Fig. 4, Appendix A"},{"comment":"The term 'Kondo cloud' denotes a nonlocal many-body correlation, but no nonlocal spin-spin correlator, Kondo length, or real-space entanglement measure is computed anywhere in the manuscript. The conclusion that the cloud remains 'largely intact' is inferred from the persistence of the central conductance peak and the absence of Floquet sidebands on the Kondo feature. Local Kondo pinning alone does not establish the integrity of the nonlocal cloud; local decoherence or an effectively local broadening mechanism could produce similar transport features. The authors should either compute a genuine nonlocal correlation function or explicitly weaken the claim to the persistence of local Kondo correlations, which is what the conductance data actually support.","section":"§IVD, Fig. 5"},{"comment":"The abstract states that the approach provides 'controlled access' to dynamical response functions and nonequilibrium transport. The benchmarking, however, is only against 2PT and GW, both of which are approximate resummations with no exact steady-state comparison; in equilibrium, the analytic benchmark of Fig. 6(b) shows that the method does not capture the Kondo scale. Thus the term 'controlled' overstates what is demonstrated. The truncation in Eq. (48) plus neglect of inter-channel mixing is a consistent approximation scheme, but its accuracy for the Kondo scale is unknown. I recommend either adding a numerically exact comparison for at least one parameter set, or tempering the 'controlled' claim to describe a systematic approximation whose quantitative accuracy is verified only in the weak-to-intermediate coupling regime.","section":"§IIIE, §IVC"}],"minor_comments":[{"comment":"There is a typographical error in the retarded Green's function: '⟨{cσ(t), c†σ′(t′)]}⟩' contains an unbalanced closing bracket. Please correct.","section":"Eq. (10)"},{"comment":"The notation δ(t′1=t′2=t1=t2) for the instantaneous vertex is nonstandard and slightly ambiguous; a standard product of delta functions would be clearer.","section":"Eq. (14)"},{"comment":"The statement that 'compared to Ref. [60] the x-axis must be scaled by a factor 1/2' is unclear. Please specify which quantity is scaled and why, or remove the remark if it is not essential.","section":"Fig. 6 caption"},{"comment":"The frequency-grid scale factor B=15 is described as heuristic. A brief convergence statement with respect to B would be helpful, since all frequency-resolved results depend on this choice.","section":"Eq. (76), §IIIG1"},{"comment":"The caption states that T_res=0.05Δ is 'still below the Kondo temperature for the shown interaction strengths.' For the largest U/Δ values, the Kondo temperature is exponentially small, so this statement may not hold; please provide the Kondo-temperature estimate used.","section":"Fig. 4 caption"}],"recommendation":"major_revision","confidential_remarks":null},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Bottom line: this is a real methods paper with an overreach in its headline physics claim. The frequency-resolved Floquet FRG is new, carefully derived, and benchmarked about as well as one can hope at this level. The Kondo-cloud-robustness claim, however, rests on the very truncation that the paper's own Appendix A shows cannot reproduce the exponential Kondo scale. Send it to review, but expect the authors to be pushed on the physical conclusion.\n\nWhat is genuinely new: they formulate the frequency-dependent vertex FRG directly in Floquet steady-state space, with channel decomposition, Katanin substitution, and hybridization flow. That is a natural extension of Refs [59,60], but the implementation is nontrivial and, as far as I know, not done before. The numerical details are transparent: causality checks pass, the frequency grid and Floquet truncation are documented, and the sFRG/rsFRG static approximations are tested against the full frequency-resolved scheme. The weak-to-intermediate coupling agreement with 2PT and GW is real, and the transport result—the photo-induced replica peak shifting from Vg = Omega to Vg = Omega + U/2—is interesting and argued carefully.\n\nNow the soft spots, in proportion. The central physical claim is that the Kondo cloud remains largely intact under strong local driving. The evidence is m* in Fig. 4 and conductance in Fig. 5. But Appendix A and Fig. 6(b) show that this same channel-decomposed one-loop FRG misses the equilibrium exponential scaling m* ~ exp(U/Delta) and overestimates m* for weak drive. The paper itself concedes 'the FRG could overestimate the suppression.' So the robustness statement is being made by an approximation that demonstrably misrepresents the very quantity being measured, and no numerically exact steady-state benchmark is provided. Also, 'Kondo cloud' is inferred from local pinning and the absence of replicas; no direct nonlocal spin-spin correlation is computed. That is a plausible inference, but it is not a measurement of the cloud. These are not hidden flaws—the authors call for exact verification—but they should be acknowledged as unproven.\n\nMinor points: the frequency-grid scale factor B in Eq. (76) is a heuristic free parameter, and the data/code are not publicly available. The self-citation overlap is heavy but not unreasonable given the method builds directly on those papers.\n\nWho is this for? People working on Floquet impurity solvers and nonequilibrium FRG. The method section will be the lasting contribution; the Kondo-cloud claim should be treated as a conjecture. A serious referee should ask for an exact steady-state benchmark in the Kondo regime (the semigroup influence matrix method they cite is a good candidate), or at least a direct nonlocal correlator, before the headline is accepted. The paper deserves peer review, not desk rejection.","headline":"A genuine new Floquet-FRG method; the method claim holds up, the Kondo-cloud-robustness headline does not—both for the same reason: the truncation misses the exponential Kondo scale.","tokens_in":28299,"tokens_out":1899,"would_cite":true,"duration_ms":20966,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Periodic driving of a quantum dot broadens the Kondo resonance inelastically, yet the nonlocal Kondo cloud stays largely intact, preserving a fraction of Kondo pinning even under a strong local Floquet drive.","keywords":["Floquet steady state","functional renormalization group","Keldysh formalism","Anderson impurity model","Kondo effect","nonequilibrium transport","driven quantum dot","spectral function"],"falsifier":"A numerically exact computation of the driven Anderson impurity steady state (e.g., using a quantum Monte Carlo or density-matrix renormalization group simulation of the infinite-time Floquet state) at U/Δ ≈ 4–6, Ω/Δ ≈ 5, and A/Δ ≈ 5, comparing the effective mass m* and the linear conductance with the FRG predictions, and checking whether the Kondo peak's Floquet sidebands are truly absent.","tokens_in":27380,"feed_emoji":"🔄","tokens_out":3186,"duration_ms":29683,"temperature":0.7,"pith_summary":"This paper introduces a functional renormalization group (FRG) framework formulated directly in the Floquet steady state, keeping the frequency dependence of the two-particle vertex. The authors argue that this gives controlled access to dynamical response functions and nonequilibrium transport in driven interacting systems. They benchmark it on the periodically driven single-impurity Anderson model against Floquet Green's function methods, finding quantitative agreement for finite-frequency observables up to intermediate interaction strengths. Their main physical conclusion is that a local periodic drive broadens the Kondo resonance through inelastic scattering, but the nonlocal Kondo cloud is largely unaffected, suppressing Floquet replicas of the Kondo peak and leading to a partial persistence of Kondo pinning.","feed_headline":"Driving broadens Kondo peak, leaves Kondo cloud intact","feed_subtitle":"Frequency-resolved renormalization group shows local periodic drive suppresses resonance replicas but holds many-body singlet.","key_machinery":"The central object is the frequency-dependent two-particle vertex in Floquet–Keldysh space, decomposed into particle-particle, particle-hole, and direct particle-hole channels. The flow equations for these channel vertices, together with the self-energy, are integrated using a hybridization-flow regulator and the Katanin substitution, yielding an RG-improved ladder resummation that goes beyond RPA while retaining frequency dependence—this is what enables the treatment of inelastic scattering and finite-frequency response.","core_discovery":"This paper claims to establish a frequency-resolved Floquet FRG scheme that retains the frequency structure of the two-particle vertex to second order in the interaction, enabling controlled access to spectral functions, effective masses, and transport in driven interacting systems. Applied to the single-impurity Anderson model with a back-gate drive, it finds that local periodic driving strongly broadens the Kondo resonance via inelastic scattering, while the many-body Kondo cloud remains largely intact. This robustness suppresses Floquet sidebands associated with the Kondo resonance and leads to a partial persistence of Kondo pinning, indicating that the nonlocal many-body singlet is resil","pith_inferences":["If the Kondo cloud is indeed left intact by local driving, nonlocal impurity-reservoir entanglement may survive even when the local resonance is strongly damped; one could test this directly by computing the impurity-reservoir spin correlation length or entanglement entropy in the driven steady state.","The suppression of Floquet sidebands on the Kondo peak implies that local Floquet engineering of the Kondo effect is limited by decoherence; globally driving the reservoirs or coupling to multiple reservoirs could behave differently, as the paper itself notes.","The exponential Kondo-scale failure of one-loop FRG suggests the magnitude of resonance suppression at large U/Δ is likely overestimated; an exact steady-state benchmark would clarify whether the partial persistence of Kondo pinning is quantitatively reliable or only qualitatively correct.","The replica-static approximation (rsFRG) may serve as a low-cost tool for driven lattice models where full frequency resolution is too expensive, extending the approach to extended systems."],"forward_implications":["The method provides a tractable route to compute dynamical response functions and conductances in driven interacting impurity systems beyond static self-energy approximations.","Static approximations to the vertex flow (sFRG) break down at intermediate interaction strengths, but a replica-resolved static approximation (rsFRG) remains close to the fully frequency-dependent FRG for static quantities.","The physical result implies that local driving acts as a decoherence source that broadens the Kondo resonance, yet nonlocal Kondo correlations are more robust than the resonance itself.","This suggests a practical workflow for driven quantum materials: simpler static approximations may suffice for static properties, while the frequency-resolved FRG is needed for accurate dynamical properties."],"fun_headline_variants":["Drive widens Kondo peak, preserves cloud","Floquet RG: Kondo peak broadens, cloud survives","Driven Anderson impurity: Kondo resilience revealed","Kondo cloud robust under periodic drive","New RG for Floquet systems shows Kondo pinning persists"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The Kondo-cloud robustness conclusion relies on a one-loop channel-decomposed FRG approximation that does not capture the exponential Kondo scale in equilibrium and may overestimate the suppression of the Kondo resonance.","fun_headline_variants_meta":{"raw":{"variants":["Drive widens Kondo peak, preserves cloud","Floquet RG: Kondo peak broadens, cloud survives","Driven Anderson impurity: Kondo resilience revealed","Kondo cloud robust under periodic drive","New RG for Floquet systems shows Kondo pinning persists"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000548,"raw_usage":{"total_tokens":2438,"prompt_tokens":709,"completion_tokens":1729,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":453,"completion_tokens_details":{"reasoning_tokens":1652}},"tokens_in":453,"tokens_out":1729,"duration_ms":10513,"temperature":1.0,"reasoning_tokens":1652,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-02T18:15:01.859457+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A numerically exact computation of the driven Anderson impurity steady state (e.g., using a quantum Monte Carlo or density-matrix renormalization group simulation of the infinite-time Floquet state) at U/Δ ≈ 4–6, Ω/Δ ≈ 5, and A/Δ ≈ 5, comparing the effective mass m* and the linear conductance with the FRG predictions, and checking whether the Kondo peak's Floquet sidebands are truly absent.","supporting_citations":[],"review_version":1}