{"id":"f2a81544-cc53-4eee-9b58-cfd02199448c","arxiv_id":"2507.01683","paper_version":1,"verdict":"ACCEPT","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"low","formal_verification":"none","parameter_count":1,"one_line_summary":"A noisy quantum channel can simulate a perfect state transfer via a quasiprobability recipe whose sampling overhead is 2/F - 1, where F is the channel's entanglement fidelity, validated on IBM quantum hardware.","lead":"This paper shows how to simulate a perfect quantum connection using a noisy real connection, by mixing the noisy operation with classical operations in a weighted random recipe. It only requires measuring one number about the connection quality, and tests on IBM quantum chips show the simulation beats using the noisy connection directly.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Calibration accuracy in the low-fidelity regime is the weakest link; Theorem 1 itself is sound.","rationale":"The central theorem is a straightforward and correct QPD identity: full twirl of C gives D_{F(C)}, Lemma 1 constructs D0 exactly, so Eq. (39) is valid and the overhead formula follows. My concern targets the experimental bridge. Fig. 6 is the paper's own validation of the one-parameter calibration; at F ≈ 0.52–0.57 the computed coefficient 1/F visibly misses the empirical optimum c_opt, while at high F it agrees. The authors ascribe this to sensitivity of 1/F to errors in F, but the same pattern appears if F drifts between calibration and execution or if the Section 3.2 estimator is biased by imperfect twirls. Eq. (55) shows the consequence: a coefficient mismatch produces a systematic bias that does not vanish as N grows. Since the paper's headline practical claim includes straightforward calibration and overhead reduction, this is the weakest load-bearing step. It does not invalidate Theorem 1, and the authors do disclose limitations, so the verdict remains unchanged; a quantitative calibration check in the low-fidelity regime would strengthen the practical claim.","tokens_in":43723,"tokens_out":19290,"duration_ms":249231,"concrete_test":"Re-run the Fig. 6 coefficient scan at F ≈ 0.53 with an independent, interleaved F(C) measurement (randomized benchmarking or process tomography) taken immediately before and after the QPD runs. If c_com still differs from c_opt by more than the statistical uncertainty of the scan, the Section 3.2 calibration is biased in the low-fidelity regime and the practical claim should be scoped accordingly.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Theorem 1 is a correct algebraic identity: full twirl of C gives D_{F(C)}, Lemma 1 constructs D0 exactly, and the overhead formula follows. The load-bearing practical step is the one-parameter calibration of Section 3.2. In the authors' own coefficient-scan validation (Fig. 6), the computed coefficient c_com = 1/F(C) matches the empirical optimum c_opt at high fidelity (F ≈ 0.87–0.91) but deviates noticeably at low fidelity (F ≈ 0.52–0.57), exactly where the QPD's sampling-overhead advantage over classical wire cutting is marginal (κ = 2/F − 1 just below 3). The paper attributes this to error magnification in the inverse map c = 1/F, but drift or a biased F estimate produces the same signature. Because the QPD estimator is unbiased only when the coefficients equal the true 1/F, any mismatch contributes the third term of Eq. (55) and biases the reported error curves. Thus the experiments do not fully validate the calibration method in the regime where its accuracy matters most, even though the central theorem remains unaffected.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The authors propose a quasiprobability decomposition (QPD) that simulates an ideal n-qubit state transfer by combining a twirled noisy channel E_E(C) with a measure-and-prepare-based depolarizing channel E_V(M). Theorem 1 establishes I = (1/F(C)) E_E(C) - (1/F(C)-1) E_V(M) for any channel C with entanglement fidelity F(C), giving sampling overhead kappa = 2/F(C) - 1. The single parameter F(C) is calibrated through the twirl-based relation F(C) = (2^n + 1) P_{0->0}/2^n - 1/2^n (Eq. 42). The paper also studies reduced twirling ensembles (Pauli mixing vs. unitary two-design) and provides an error analysis in Appendix E that separates statistical sampling error from systematic bias. Experiments on three IBM devices emulate a teleportation channel degraded by controlled SWAP operations and report O(1/sqrt(N)) error scaling, lower errors than direct teleportation, and coefficient scans matching the calibration at high fidelity.","tokens_in":43933,"tokens_out":4914,"duration_ms":60538,"significance":"If the result holds, it is a genuinely practical advance: a noisy interconnect can be used as a resource to simulate a perfect state transfer with an overhead that decreases as interconnect quality improves, and the calibration requires only a single scalar parameter. The paper's strengths are substantial: Theorem 1 and Lemma 1 have complete proofs in the appendices, the error analysis in Appendix E is careful (including the coherent-error bound of Eq. 56), the code and data are publicly available, and the hardware experiments on three devices test the predicted scaling rather than curve-fitting it. The experimental evidence that the unitary-two-design QPD beats direct teleportation even on real devices is particularly valuable. The main caveat is that coefficient validation at low fidelity (F ~ 0.52-0.57) is less conclusive than at high fidelity, but this does not affect the correctness of the central theorem.","major_comments":[],"minor_comments":[{"comment":"The discrepancy between the computed coefficient c_com and the empirical optimum c_opt in the low-fidelity panels (F ~ 0.52-0.57) is attributed to inverse error magnification, but channel drift or a biased F estimate would produce the same signature; because the third term of Eq. (55) is proportional to |p/p_tilde - 1|, a sentence quantifying the expected bias at these fidelities, or explicitly restricting the validation claim to F > 0.65, would strengthen the calibration section.","section":"Section 4.5.1, Figure 6"},{"comment":"The text contains a typo: 'representaion' should be 'representation'.","section":"Section 2.2"},{"comment":"The text contains a typo: 'fidelites' should be 'fidelities'.","section":"Section 4.3.4"},{"comment":"The phrase 'twirling any any Pauli channel' contains a duplicated 'any'.","section":"Section 2.5"},{"comment":"The description of the core set of 15 circuits would benefit from explicitly stating whether the circuit set includes the resource-state preparation and SWAP degradation operations, since the text says the set is executed 'once for each combination of noise level, one of 25 Haar-random initial states, and observables X and Z'.","section":"Section 4.3.5"}],"recommendation":"minor_revision","confidential_remarks":"The paper fits the journal's scope well and the authors are appropriately self-critical about their emulation-based hardware experiments. The low-fidelity calibration concern raised in the reviewing process is real but localized; I would not block acceptance on it, though a short discussion of drift or a restricted validation claim would improve the paper."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nThis paper gives you a one-parameter QPD for simulating ideal state transfer through any noisy channel, with sampling overhead 2/F(C)-1. The core identity is essentially a rearrangement of the depolarizing channel definition, and Pednault already used a similar QPD for wire cutting. But the generalization to arbitrary mixed channels with a single measured parameter, the minimal ancilla-free D0 construction (Lemma 1), and the first hardware test on three IBM devices are genuinely new. The proofs in the appendices are complete, and the error analysis, including the Pauli-mixing bound in Eq. (56), is careful.\n\nThey ship code and data, and the experiments match the predicted O(1/sqrt(N)) scaling. The two-design QPD beats direct teleportation, which is the expected behavior if the theory is right.\n\nThe soft spot is the calibration method in Section 3.2. It reuses a single measurement of F(C) and assumes the channel is stable. In the low-fidelity regime, Figure 6 shows the computed coefficient c_com deviating from the empirical optimum c_opt. The paper attributes this to error magnification in 1/F, which is plausible, but drift or a biased F estimate would produce the same signature. Since the QPD estimator is unbiased only when the coefficients are exactly 1/F, any mismatch biases the error curves. So the experiments do not fully validate the calibration in the regime where it matters most.\n\nThe other limitations are acknowledged and acceptable: the 'interconnect' is emulated inside a single device with SWAPs, not a real link; only single-qubit transfers are tested; and there is no direct comparison to classical wire cutting. None of this undermines Theorem 1, which is exact. The issue is purely about how reliably you can estimate F in practice, and the authors are candid about the scope.\n\nThis is a solid, incremental contribution for people working on circuit cutting and distributed NISQ computing. It deserves a serious referee and probably acceptance after minor revisions. I would ask them to tighten the calibration discussion, but I would not hold the paper up over it.","headline":"Simple, correct QPD with solid proofs and honest hardware tests; calibration at low fidelity is the real weak point, not the math.","tokens_in":44475,"tokens_out":3563,"would_cite":false,"duration_ms":37244,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["81P68","81P45"],"pacs":["03.67.-a","03.67.Hk","03.67.Lx"],"model":"deepseek-v4-flash","headline":"A noisy quantum interconnect can simulate a perfect state transfer.","keywords":["quasiprobability decomposition","quantum state transfer","sampling overhead","wire cutting","entanglement fidelity","noisy interconnects","channel twirling","distributed quantum computing"],"falsifier":"For a channel with known coherent errors, run the Pauli-mixing QPD at increasing shot counts and check whether the estimation error plateaus at a value bounded by the sum of the off-diagonal χ-matrix entries; observing no plateau, or a plateau independent of coherent error strength, would contradict Equation (56).","tokens_in":43519,"feed_emoji":"🔗","tokens_out":5939,"duration_ms":66743,"temperature":0.7,"pith_summary":"This paper claims that a noisy quantum interconnect can be turned into a virtual perfect one by expressing the ideal state-transfer channel as a quasiprobability decomposition (QPD) built from the noisy channel and a measure-and-prepare circuit. The decomposition is governed by a single number: the entanglement fidelity F(C) of the interconnect. Sampling from the QPD simulates the ideal transfer with an overhead κ = 2/F(C) − 1, which falls as the interconnect improves and beats pure wire cutting whenever F(C) > 2^(−n). The authors prove this identity, give a one-time calibration for F(C), test it on three superconducting devices, and find that the two-design version of the QPD achieves lower estimation error than direct teleportation over the same noisy channel.","feed_headline":"Noisy interconnects simulate perfect state transfer","feed_subtitle":"A two-term decomposition cuts sampling overhead and beats direct teleportation on real hardware.","key_machinery":"The carrying object is the depolarizing-channel decomposition $I^{\\otimes n} = \\frac{1}{p} D_p - \\left(\\frac{1}{p} - 1\\right) D_0$ with $p = F(C)$, combined with channel twirling to realize $D_p$ from the noisy channel C and Lemma 1's construction of $D_0$ from $2^n + 1$ measure-and-prepare circuits. This identity turns one scalar channel-quality parameter into the QPD coefficients, and the minimal circuit construction keeps the number of distinct circuits small. The same decomposition is then adapted to wire cutting with shared non-maximally entangled states, recovering the optimal overhead $2/f(\\rho) - 1$ while allowing the resource state to be mixed rather than pure.","core_discovery":"The central result is Theorem 1: for any n-qubit channel C with entanglement fidelity F(C), the identity channel admits the QPD $I^{\\otimes n} = \\frac{1}{F(C)} \\mathcal{E}_E(C) - \\left(\\frac{1}{F(C)} - 1\\right) \\mathcal{E}_V(M)$, where $\\mathcal{E}_E(C)$ is a channel twirl that turns C into a depolarizing channel, and $\\mathcal{E}_V(M)$ is a twirled measure-and-prepare channel with zero entanglement fidelity. The corresponding sampling overhead is $\\kappa = 2/F(C) - 1$, so better interconnects require fewer samples. The paper also shows that the zero-fidelity channel $D_0$ can be realized with the minimal number of $2^n + 1$ measure-and-prepare circuits, and that the QPD's single parameter can be calibrated by measuring the probability that the twirled channel preserves the all-zero state. Hardware experiments confirm that this QPD reduces estimation error relative to direct use of the noisy channel, and that a full two-design is robust to coherent errors while a smaller Pauli-mixing ensemble works well when coherent errors are small.","pith_inferences":["Editorial extension: the same two-term QPD template could be applied to simulate other distributed operations, such as multi-qubit gates, with the sampling overhead governed by an analogous fidelity parameter rather than by full process tomography.","Editorial extension: because the QPD depends on only one scalar, a quantum network could in principle route state transfers through whichever interconnect has the highest current F(C), using the calibration method as a lightweight link-quality probe.","Editorial extension: the hardware experiments emulate an inter-device interconnect with teleportation inside a single device; the direct next test is to run the QPD across genuinely remote processors and check that the overhead reduction survives inter-device synchronization and communication latency."],"forward_implications":["If F(C) > 2^(−n), the QPD has lower sampling overhead than classical-communication-only wire cutting, and the overhead approaches 1 as the interconnect approaches perfection.","On channels with negligible coherent errors, the three-circuit Pauli-mixing ensemble performs as well as the twelve-circuit unitary two-design, so the number of distinct circuits can be reduced without accuracy loss.","A single calibration of F(C), measured through the all-zero preservation probability, suffices to set all QPD coefficients for subsequent runs.","The simulated state transfer achieves higher effective fidelity than direct transfer over the same noisy interconnect, even when approximations that reduce the number of circuit variants are used."],"supporting_citations":[{"why":"Supplies the QPD sampling estimator and the O(κ²/ε²) shot scaling that motivates minimizing κ.","marker":"[13]"},{"why":"Gives the optimal wire-cutting overhead γ = 2^(n+1) − 1 that Theorem 1's overhead is compared against.","marker":"[15]"},{"why":"Provides the diagonalizing unitaries V_j and the minimal-count bound for ancilla-free measure-and-prepare realizations of D0.","marker":"[17]"},{"why":"Introduces wire cutting with non-maximally entangled states, which this work extends from pure to mixed resource states.","marker":"[18]"},{"why":"Gives the joint-wire-cutting QPD for pure NME states and the fidelity-of-distillation overhead formula used in Section 3.3.","marker":"[19]"},{"why":"Establishes that full channel twirl produces a depolarizing channel with unchanged entanglement fidelity, the basis for constructing D_p from C.","marker":"[37]"},{"why":"Supplies the incoherent-plus-coherent error model used to benchmark the QPD variants numerically.","marker":"[50]"},{"why":"Provides the contemporary superconducting devices and calibration data used for the hardware validation.","marker":"[51]"}],"fun_headline_variants":["Noisy links fake perfect state transfer with fewer samples","Two-term trick turns noisy interconnects into ideal transfer","Cut wires, not fidelity: QPD beats direct transfer","Snap noisy interconnects into perfect teleportation"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The central premise is that one calibration of the interconnect's entanglement fidelity stays valid over the QPD run, because drifting noise would make the QPD coefficients wrong and add systematic bias.","fun_headline_variants_meta":{"raw":{"variants":["Noisy links fake perfect state transfer with fewer samples","Two-term trick turns noisy interconnects into ideal transfer","Cut wires, not fidelity: QPD beats direct transfer","Snap noisy interconnects into perfect teleportation"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000809,"raw_usage":{"total_tokens":3574,"prompt_tokens":996,"completion_tokens":2578,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":612,"completion_tokens_details":{"reasoning_tokens":2512}},"tokens_in":612,"tokens_out":2578,"duration_ms":22377,"temperature":1.0,"reasoning_tokens":2512,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T20:45:47.085265+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"For a channel with known coherent errors, run the Pauli-mixing QPD at increasing shot counts and check whether the estimation error plateaus at a value bounded by the sum of the off-diagonal χ-matrix entries; observing no plateau, or a plateau independent of coherent error strength, would contradict Equation (56).","supporting_citations":[{"cited_title":"Error mitigation for short-depth quantum circuits","cited_arxiv_id":null,"evidence_quote":"Supplies the QPD sampling estimator and the O(κ²/ε²) shot scaling that motivates minimizing κ."},{"cited_title":"Doubly optimal parallel wire cut- ting without ancilla qubits","cited_arxiv_id":null,"evidence_quote":"Provides the diagonalizing unitaries V_j and the minimal-count bound for ancilla-free measure-and-prepare realizations of D0."},{"cited_title":"Cutting a wire with non-maximally entangled states","cited_arxiv_id":null,"evidence_quote":"Introduces wire cutting with non-maximally entangled states, which this work extends from pure to mixed resource states."},{"cited_title":"Joint wire cutting with non-maximally entan- gled states","cited_arxiv_id":null,"evidence_quote":"Gives the joint-wire-cutting QPD for pure NME states and the fidelity-of-distillation overhead formula used in Section 3.3."},{"cited_title":"General teleportation chan- nel, singlet fraction, and quasidistillation","cited_arxiv_id":null,"evidence_quote":"Establishes that full channel twirl produces a depolarizing channel with unchanged entanglement fidelity, the basis for constructing D_p from C."},{"cited_title":"In-situ characterization of quantum devices with error correction","cited_arxiv_id":"1405.5656","evidence_quote":"Supplies the incoherent-plus-coherent error model used to benchmark the QPD variants numerically."},{"cited_title":"IBM Quantum De- vices","cited_arxiv_id":null,"evidence_quote":"Provides the contemporary superconducting devices and calibration data used for the hardware validation."}],"review_version":1}