{"id":"6ba528fc-d08c-49cc-b78b-79f4dd09ebac","arxiv_id":"2607.21360","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"A Floquet reservoir-engineering protocol stabilizes a remote logical Bell state and distills weak bath entanglement into protected qubits, turning the loss error scaling from ε^(1/2) to ε.","lead":"This paper proposes a protocol that alternates engineered dissipation with fast quantum gates to stabilize remote entangled logical qubits without measurements or classical communication. It acts like an autonomous entanglement distillation engine and claims improved protection against photon loss compared to direct physical-qubit entanglement.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Uniqueness of the steady state in Eq. (5) is asserted for all τ_stb, τ_conv but not proven; footnote [49] admits the analytic argument covers only τ_stb ≫ τ_ent, and the channel may have other fixed points at finite intervals, so the central stabilization claim is not yet established.","rationale":"The reader's conditional verdict is well matched to the strongest unresolved issue: the uniqueness of the steady state asserted in Eq. (5). The reader's weakest_assumption emphasized the external hardware premise (a pure entangled reservoir with non-invariant even-parity subspace), which is important but not the most load-bearing internal gap. The more direct problem is that the paper itself, in footnote [49], admits the uniqueness claim is not proven for arbitrary τ_stb and τ_conv, and the analytic rate-equation derivation only covers the τ_stb → ∞ limit. That missing proof is load-bearing because the protocol's usefulness—including the linear-in-ε loss protection—depends on repeated application of E converging to the logical Bell state from arbitrary initial states. The proposed concrete test—spectral analysis of the finite-dimensional channel—can settle whether additional fixed points exist at finite times. I do not see a reason to change the reader's verdict: the paper is plausible and the analytic TMS p_conv derivation is a genuine strength, but the central uniqueness theorem should be proven or explicitly qualified. Thus 'UNCHANGED' (keep CONDITIONAL) is appropriate.","tokens_in":25563,"tokens_out":7874,"duration_ms":83360,"concrete_test":"Build the Liouville superoperator of E = E_ZZ ∘ E_XX on the 16-dimensional joint Hilbert space (256×256 density-matrix representation) for the TMS dissipator Eq. (S13). Use exact matrix exponentials to form M(τ) at finite τ, and compose with the four controlled-unitary maps. Compute the full spectrum on a grid of (γτ_stb, γτ_conv) values including 0.1, 1, 5, 10 and several squeezing parameters r. Check whether any eigenvector with |λ| = 1 lies outside the one-dimensional span of |Φ+_L>⊗|ψ^c_ss>. Repeat for the Rabi dissipator Eq. (S15). If any such eigenvector exists, Eq. (5)'s uniqueness claim is false; if none exists on a dense grid, the claim gains numerical support but still lacks an analytic proof.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is Eq. (5): 'no matter what the choice of time intervals τ_stb, τ_conv ... the composite channel E has a unique pure steady state' |Φ+_L>⊗|ψ^c_ss>. The paper's own footnote [49] explicitly disclaims an analytic proof for arbitrary parameters: 'we do not have an analytic proof of uniqueness for arbitrary parameter regime; ... Our conclusion relies on numerics beyond this scope.' The analytic derivation in SM SI.C and Eqs. (8)-(9) assumes τ_stb → ∞ and tracks only the parity-diagonal populations (p_e, p_o). It does not constrain off-diagonal logical coherences, nor does it address finite-τ_stb contamination that could open additional invariant subspaces. In fact, the literal claim fails at τ_stb = τ_conv = 0: E reduces to a unitary controlled-gate map with many steady states. Since the loss-protection advantage in Fig. 3 is computed as a steady state of E (SM SV.B), an unproven non-uniqueness—or an extremely slow approach to the target—undermines the headline result. This is a load-bearing gap, not a mere technicality.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper introduces a Floquet reservoir engineering protocol for stabilizing a maximally entangled Bell state between two remote logical qubits. The protocol repeatedly applies two subroutines, E_ZZ and E_XX, each consisting of entangling dissipation on communication qubits interleaved with fast local controlled-logical gates. The authors argue that the composite channel E has the unique pure steady state |Φ_L^+>⊗|ψ_c^ss> for essentially arbitrary timing parameters, provide analytic rate equations for the stabilizer correlators in the τ_stb ≫ τ_ent limit, derive a closed-form conversion probability p_conv for the two-mode-squeezing reservoir, and show numerically and via perturbation theory that waveguide loss yields a linear (∝ε) steady-state concurrence error, outperforming the ε^{1/2} scaling of direct communication-qubit stabilization. They also map the protocol to a cat-qubit/transmon architecture.","tokens_in":25977,"tokens_out":14679,"duration_ms":151916,"significance":"If the central claims hold, this is a valuable proposal: it overcomes known time-entanglement tradeoffs, provides autonomous entanglement distillation, and offers a concrete path to remote logical entanglement in modular superconducting systems. The paper contains several strong elements: (i) the analytic derivation of p_conv for the TMS reservoir via quantum trajectories (SM SIII), (ii) the transparent two-correlator rate equations (Eqs. (8)-(9)), (iii) numerical verification of the convergence and loss scaling, and (iv) a detailed, bias-preserving experimental implementation for cat qubits. The main weakness is that the headline uniqueness claim is not proven in the claimed generality and, as stated, is false at τ_conv = 0; this is acknowledged in footnote [49] but the unqualified statement in Eq. (5) remains a load-bearing issue. A second concern is that the loss-optimal operating point has vanishing stabilization rate, so the reported steady-state advantage may not translate to finite-time experiments.","major_comments":[{"comment":"The claim that E has a unique pure steady state for 'no matter what the choice of time intervals τ_stb, τ_conv' is not established and is literally false at τ_stb=τ_conv=0, where E reduces to a unitary (products of C_Z and C_X) with many steady states. At τ_conv=0 with finite τ_stb, p_conv=0 and the rate equations (8)-(9) give no convergence, so additional fixed points exist. The analytic treatment in SI.C assumes τ_stb→∞ and tracks only parity populations; it does not constrain off-diagonal logical coherences or finite-τ_stb corrections. Since the loss-protection results (Fig. 3, SM SV.B) compute a steady state of E, an unproven non-uniqueness could make the reported concurrence initial-state dependent. Please either prove uniqueness under precise conditions (e.g., τ_stb sufficiently large and p_conv>0) or qualify the claim and demonstrate numerically that the same steady state is reach","section":"Stabilization mechanism, Eq. (5) and footnote [49]; SM SI.C"},{"comment":"The optimal steady-state concurrence under loss is found at r→0 (TMS), where p_conv∝r^2 and the stabilization rate Γ = -ln(1-p_conv) vanishes. Thus the reported linear scaling 1-C≲16ε is an infinite-time steady-state result, and the time needed to reach it diverges. The main-text discussion of Fig. 3 does not mention this, and the 'advantage' over communication-qubit-only stabilization may not persist at any finite run time. Please add a finite-time analysis or a rate constraint to the loss optimization, or clearly state in the main text that the advantage is only asymptotic in the number of cycles.","section":"End Matter Eq. (A.12) and SM SV.C"}],"minor_comments":[{"comment":"The unqualified wording 'no matter what the choice of time intervals' should be replaced by a statement that includes the conditions under which uniqueness is actually shown or expected.","section":"Eq. (5) / Footnote [49]"},{"comment":"The closed form for p_conv is a highlight, but the notation γ' = γ cosh 2r is introduced only in the text; define it next to the equation.","section":"SM Eq. (S22)"},{"comment":"The figure caption and text should state whether the entanglement parameter r (or Ω/γ) is also optimized at each ε. SM SV.C suggests r→0 is optimal, which is not visible from Fig. 3.","section":"Fig. 3"},{"comment":"The saturation of Γ with τ_stb is shown in Fig. S1, but the analytic equation for the residual finite-τ_stb correction is not given; a bound or explicit exponential form would strengthen the argument.","section":"SM SIV"},{"comment":"Several self-references to the supplemental material use placeholder [53]; include the actual link or reference. Also, the notation for the controlled gates alternates between C_Z and CZ; pick one style.","section":"General"}],"recommendation":"major_revision","confidential_remarks":"The paper is interesting and the core mechanism is probably right, but the uniqueness claim is overstated and the vanishing-rate issue tempers the practical significance of the loss advantage. I would support publication after the authors either prove uniqueness under stated conditions or carefully restrict all claims to the regime they can support, and after they address the finite-time tradeoff. The analytic p_conv derivation and the cat-qubit implementation are strong assets."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Let me give you the short version. Yao and Clerk have a genuinely new construction: interleave continuous engineered dissipation on communication qubits with fast local controlled-logical gates, so the bath's weak entanglement gets distilled into a remote logical Bell state. No measurements, no classical communication. The mechanism is clear, the TMS analytic conversion probability (Eq. S22) is a real derivation—quantum-jump counting with a weak symmetry—and the numerics back the rate equations. The loss analysis is also sound in structure: they separate false-positive errors from direct corruption and derive the linear-in-ε steady-state formula.\n\nThe soft spots are where the claims outpace the proof. The 'no matter what the choice of τ_stb, τ_conv' uniqueness statement in Eq. (5) is not actually proven; footnote [49] says the analytic argument only covers τ_stb ≫ τ_ent, and for general parameters they rely on numerics. At the degenerate point τ_stb = τ_conv = 0 the channel is just a unitary gate map with many steady states, so the sentence as written is literally false. That matters because the loss-protection curves in Fig. 3 are computed as steady states of the same channel; if there are multiple fixed points at finite times, they might be optimizing toward the wrong one. I don't think the construction is wrong—the numerics look plausible—but this is a load-bearing gap and the authors should either prove uniqueness in the relevant regime or qualify the claim.\n\nThe second issue is the time cost. The linear-in-ε advantage is real, but the optimal operating point pushes r→0 (SM SV.C), where the stabilization rate vanishes. The main text shows the error scaling but not the number of cycles needed. The End Matter has a run-time figure, but the connection to the loss comparison is left implicit. A reader comparing protocols needs to know that the ε advantage comes at an exponentially slow convergence cost.\n\nThe hardware premise—a reservoir with a unique pure entangled steady state and even parity not invariant—is stated cleanly and satisfied by both examples. That's fine.\n\nBottom line: the protocol is clever and the analytics are decent, but the paper oversells the uniqueness and hides the convergence tradeoff in the loss comparison. It deserves a serious referee; I'd send it out, but ask for a revised main text that either proves uniqueness in the relevant parameter regime or clearly flags the numerical reliance, and that states the time-cost tradeoff alongside the loss curves. If the authors fix those, it's a strong paper.","headline":"A new Floquet dissipative stabilization scheme for remote logical Bell states—clever and mostly sound, but the headline uniqueness claim is weaker than stated and the loss advantage hides a slow-convergence catch.","tokens_in":26368,"tokens_out":3641,"would_cite":true,"duration_ms":38079,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper introduces a Floquet reservoir-engineering protocol that stabilizes a remote logical Bell state deterministically, with no measurements or classical communication.","keywords":["Floquet reservoir engineering","remote entanglement","logical qubits","dissipative stabilization","entanglement distillation","waveguide loss","cat qubits","autonomous feedback"],"falsifier":"Directly compute the steady state of E = E_ZZ ∘ E_XX for a specific reservoir (e.g., the TMS dissipator) at finite τ_stb and τ_conv where the analytic proof does not apply; if the fixed point's logical concurrence is below 1, or if under controlled waveguide loss ε the steady-state concurrence error scales as ε^{1/2} rather than ε, the central claims fail. The paper itself notes that uniqueness is proven analytically only in the τ_stb ≫ τ_ent limit and relies on numerics otherwise, so an independent scan across the (τ_stb, τ_conv, r) parameter space would settle whether the claimed unique pure","tokens_in":25465,"feed_emoji":"🔗","tokens_out":7401,"duration_ms":70013,"temperature":0.7,"pith_summary":"This paper aims to establish that a time-periodic sequence of local unitary gates interleaved with a continuously running entangling reservoir can stabilize a maximally entangled Bell state between two remote logical qubits. The central claim is that the composite Floquet channel has a unique pure steady state—the logical Bell state tensored with the reservoir's entangled communication-qubit state—independently of the chosen gate timings, as long as the reservoir possesses a nonzero entangled steady state. Because the protocol needs no measurements or classical communication, it acts as an autonomous distillation engine, concentrating weak bath-induced entanglement into a protected logical pair. The paper further claims that under waveguide loss the stabilized logical concurrence error scales linearly in the loss, whereas stabilizing the communication qubits directly gives a square-root scaling. A concrete superconducting-circuit implementation using cat qubits and transmons is analyzed.","feed_headline":"No measurements needed to stabilize remote logical Bell states","feed_subtitle":"Periodic local gates distill weak bath entanglement into a protected Bell pair, giving ε rather than √ε loss scaling.","key_machinery":"The central object is the composite Floquet channel E = E_ZZ ∘ E_XX, where each subroutine is a composition of two finite-time dissipation channels M(τ_stb) and M(τ_conv), generated by a time-independent entangling Lindbladian L_ent, and two fast controlled-logical gates (controlled-Z in E_ZZ, controlled-X in E_XX). The key quantity is the conversion probability p_conv(τ_conv): the probability that the dissipation segment M(τ_conv) turns the parity-encoded communication state |ψ_det> into an odd-parity communication state, making the logical parity locally readable. The bath's unique pure dark state |ψ_ss> supplies the resource, and the non-invariance of the even-parity subspace guarantees p","core_discovery":"The paper's central discovery is that a time-periodic, non-unitary channel E = E_ZZ ∘ E_XX, built from two finite-time dissipation segments M(τ)=exp(τ L_ent) on a pair of communication qubits and two node-local controlled-logical gates per subroutine, has the product state |Φ_L^+> ⊗ |ψ_ss> as its unique pure steady state whenever the entangling Lindbladian L_ent has a unique pure entangled dark state |ψ_ss> = u|0c_A 0c_B> + v|1c_A 1c_B> with |u|,|v|>0 and the even-parity subspace is not invariant. Each E_ZZ subroutine uses the dissipation twice: first to refresh the communication qubits into |ψ_ss>, then, after a controlled-Z gate has encoded the logical ZZ parity into a global phase, to con","pith_inferences":["Beyond the paper: the p_conv>0 condition suggests a testable trade-off—weakly entangled reservoirs with slow stabilization may be optimal under loss, because the error coefficient c grows with entanglement while p_conv grows only quadratically; the paper notes this at r→0, and one could verify it for the Rabi scheme as well.","Beyond the paper: the linear loss scaling should persist for other logical encodings as long as the local controlled-logical gates are bias-preserving; the paper analyzes cat qubits, but the black-box argument suggests a broader pattern that could be checked in GKP or surface-code simulations.","Beyond the paper: if p_conv is small, the number of cycles to steady state grows as 1/p_conv, so in practice the effective stabilization rate may be limited by logical memory errors during long preparation times; this suggests an optimal τ_stb that balances idling error against conversion rate, an optimization the paper's cat-qubit analysis hints at."],"forward_implications":["If the central claim holds, any pair of remote communication qubits that can be dissipatively entangled can be upgraded to a protected logical Bell pair using only local controlled-logical gates, independent of the specific logical encoding.","The protocol removes the time-entanglement tradeoff that limits time-independent dissipative entanglement: the logical steady state reaches unit fidelity for any p_conv>0, not only in the limit of maximal bath entanglement.","Under waveguide loss, the logical concurrence error scales as O(ε) instead of O(ε^{1/2}); the paper identifies an operating point at weak bath squeezing where the error floor is about 16ε for the TMS reservoir.","The scheme runs autonomously, with no measurements or classical communication, a practical advantage for remote nodes without a fast classical link.","Because the construction treats the logical encoding as a black box, it extends naturally to cat, GKP, and surface-code logical qubits, as demonstrated explicitly for a cat-transmon superconducting circuit."],"fun_headline_variants":["Floquet dissipation stabilizes remote logical Bell pairs autonomously","Autonomous entanglement distillation with periodic dissipation","Floquet engineering protects remote logical Bell states from loss","Periodic dissipation without measurement yields protected logical entanglement","Interleaved dissipation and gates distill remote logical Bell states"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The protocol collapses if the physical entangling reservoir does not possess a unique pure entangled steady state |ψ_ss> with |u|,|v|>0 and does not break the even-parity subspace, because then the parity-encoding map and the conversion probability p_conv that drive the feedback loop cannot be established.","fun_headline_variants_meta":{"raw":{"variants":["Floquet dissipation stabilizes remote logical Bell pairs autonomously","Autonomous entanglement distillation with periodic dissipation","Floquet engineering protects remote logical Bell states from loss","Periodic dissipation without measurement yields protected logical entanglement","Interleaved dissipation and gates distill remote logical Bell states"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000825,"raw_usage":{"total_tokens":3420,"prompt_tokens":696,"completion_tokens":2724,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":440,"completion_tokens_details":{"reasoning_tokens":2658}},"tokens_in":440,"tokens_out":2724,"duration_ms":18132,"temperature":1.0,"reasoning_tokens":2658,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-01T07:39:37.480061+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Directly compute the steady state of E = E_ZZ ∘ E_XX for a specific reservoir (e.g., the TMS dissipator) at finite τ_stb and τ_conv where the analytic proof does not apply; if the fixed point's logical concurrence is below 1, or if under controlled waveguide loss ε the steady-state concurrence error scales as ε^{1/2} rather than ε, the central claims fail. The paper itself notes that uniqueness is proven analytically only in the τ_stb ≫ τ_ent limit and relies on numerics otherwise, so an independent scan across the (τ_stb, τ_conv, r) parameter space would settle whether the claimed unique pure","supporting_citations":[],"review_version":1}