{"id":"37e50eaf-8ca8-47ba-b023-fe1c9f869aa1","arxiv_id":"2607.08967","paper_version":1,"verdict":"ACCEPT","confidence":"HIGH","novelty_score":5.5,"correctness_risk":"low","formal_verification":"none","parameter_count":4,"one_line_summary":"PPT criterion plus overlapping tomography maps pairwise spin entanglement across quantum phase transitions on up to 20-qubit hardware, recovering peaks at criticality after error mitigation.","lead":"Researchers used the PPT entanglement witness plus overlapping tomography to map all two-spin entanglement in critical spin-chain states prepared on a 20-qubit IBM processor. The method works for noisy mixed states and shows clear entanglement peaks at quantum phase transitions, offering a practical benchmark for quantum simulators.","discovery_kind":"new_application","skeptic_critique":{"model":"grok-4.5","headline":"Residual bias after partial-fold ZNE + M3, combined with sub-unity ansatz fidelity, remains the softest point for the weaker (next-nearest) λ_min signals that the claim needs.","rationale":"The reader correctly isolates the only material soft spot: whether the limited-depth brick-wall ansatz plus the chosen mitigation pipeline recovers unbiased two-spin RDMs near criticality. The paper’s own data (high classical fidelities, quantitative post-mitigation agreement with independent MPS benchmarks for both energy and λ_min, bootstrap significance, and the expected peak of |λ_min| at the TFIM and XXZ critical points) already supply direct empirical support that residual bias does not flip the reported signs for the demonstrated system sizes. No internal inconsistency, circularity, or mathematical error is present; PPT for two qubits is necessary and sufficient, QOT scaling is correct, and the multipartite caveat is acknowledged. The concern therefore remains a standard NISQ caveat rather than a reason to downgrade the verdict. ACCEPT stands; the usual request for public circuits and raw counts would further harden the result.","tokens_in":22243,"tokens_out":622,"duration_ms":30021,"concrete_test":"Re-process the N=12 XXZ (Δ=−0.68) shot data: reconstruct every individual d=2 ρ_AB without translational averaging, recompute the 1000-sample bootstrap distribution of λ_min for each pair, and test whether a clear majority remain >1 SE below zero. Separately replace partial-fold ZNE by a second mitigation method (e.g., no ZNE or PEC) on the same correlators; if the mean λ_min for d=2 shifts positive by more than the reported SE or loses statistical significance, the next-NN portion of the claim is undermined.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The strongest claim requires that the hardware-reconstructed two-spin RDMs, after M3 readout mitigation and partial-fold ZNE, recover the ideal ground-state PPT spectrum (including the weaker next-nearest-neighbor negativity visible in the XXZ gapless phase, Figs. 2d and 4c) without residual bias large enough to flip the sign of λ_min. Table I reports ansatz fidelities of 0.982–0.998; Sec. III B 2 and III C 1 explicitly note that ZNE extrapolations are occasionally unstable with large uncertainty. Because λ_min is obtained by nonlinear diagonalization of the partial transpose of a correlator-reconstructed ρ_AB, even modest residual coherent or non-Markovian bias that survives translational averaging can push a near-zero eigenvalue across the PPT threshold. Unmitigated data already lose the next-NN signal (Fig. 2b), so the post-mitigation recovery is load-bearing for the “complete map” and criticality-peak statements.","agreement_with_reader":"agree"},"referee_report":{"model":"grok-4.5","summary":"The manuscript introduces the positive partial transpose (PPT) criterion, combined with quantum overlapping tomography, as a scalable witness of two-spin entanglement for quantum-critical states prepared on noisy hardware. Using variational brick-wall circuits (optimized for fidelity to DMRG/MPS ground states) they prepare TFIM and XXZ chains of up to 20 qubits on ibm_boston, reconstruct all two-qubit reduced density matrices, apply M3 readout mitigation plus partial-fold zero-noise extrapolation, and extract the minimum eigenvalue λ_min of the partial transpose. Negative λ_min values (with bootstrap uncertainties) peak near the critical points, recover nearest- and (for XXZ) next-nearest-neighbor entanglement that matches MPS benchmarks after mitigation, and distinguish quantum from classical long-range correlations.","tokens_in":22525,"tokens_out":1014,"duration_ms":23906,"significance":"If the results hold, the work supplies a practical, model-independent, mixed-state-compatible entanglement witness that is far cheaper than entanglement entropy or controlled-SWAP protocols and is therefore well-suited both for NISQ benchmarking and for condensed-matter simulations on near-term devices. The direct hardware demonstration (20 k shots, side-by-side heat-maps versus MPS, quantitative energy and correlation agreement after mitigation, bootstrap error bars) is a concrete strength; the pipeline is immediately extensible to finite temperature and multipartite witnesses. These features make the paper a useful methodological contribution at the intersection of quantum information and quantum materials.","major_comments":[{"comment":"Sec. III B 2 and III C 1 explicitly note that partial-fold ZNE extrapolations are occasionally unstable and produce large uncertainties. Because the weaker next-nearest-neighbor negativity in the XXZ gapless phase (Figs. 2d, 4c) is recovered only after mitigation and is load-bearing for the claim of a “complete map” of two-spin entanglement, residual coherent or non-Markovian bias that survives translational averaging could still flip the sign of a near-zero λ_min. A quantitative bound on residual systematic error (e.g., fraction of unstable fits, comparison against an independent noise model, or additional intermediate scale factors) should be supplied so that the statistical significance of the NNN signal can be assessed.","section":"Sec. III B 2, III C 1; Figs. 2d, 4c"},{"comment":"Table I reports ansatz fidelities of 0.982–0.998, with deeper circuits required near criticality (Fig. 8). The hardware λ_min is compared to the exact MPS ground state, yet the prepared state is only approximately the ground state. A short analysis of how the residual variational error propagates into the two-spin RDMs (and therefore into λ_min) would clarify whether the observed peak is free of preparation bias, especially for the XXZ model where fidelity is lowest.","section":"Table I, Sec. II D 4, Fig. 8"}],"minor_comments":[{"comment":"Conclusion contains the typo “simualtions”; several section headings have stray spaces (“ENT ANGLEMENT”, “T ranspose”). A global proof-read would remove these.","section":"Conclusion, Sec. II headings"},{"comment":"Fig. 1(e) caption and main text both describe the QOT measurement settings; a single concise statement of the O(log N) scaling would avoid repetition.","section":"Fig. 1(e), Sec. II C"},{"comment":"The bootstrap procedure (1000 resamples) is described clearly, yet the precise definition of “one standard error below zero” as the significance threshold could be stated once in the methods for reproducibility.","section":"Sec. II C"},{"comment":"Appendix A gives the initial linear schedule for the variational angles; a short remark on whether the final optimized angles remain close to that schedule (or deviate strongly near criticality) would help readers assess trainability.","section":"Appendix A"}],"recommendation":"minor_revision","confidential_remarks":"The work is a solid experimental demonstration rather than a deep theoretical advance; it is appropriate for a quant-ph or quantum-technology venue. The residual-bias concern for the weaker NNN signals is real but addressable with modest additional analysis, so minor revision is sufficient. No novelty or citation issues noted."},"author_rebuttal":null,"desk_editor":{"model":"grok-4.5","letter":"This is a clean experimental paper that does exactly what the abstract claims. They take the Peres-Horodecki PPT criterion, couple it to Cotler-Wilczek overlapping tomography, prepare TFIM and XXZ ground states with a brick-wall variational ansatz (fidelities 0.98-0.99), and produce the first complete pairwise entanglement maps on IBM hardware for N=12 and 20. After M3 readout mitigation and partial-fold ZNE they recover nearest-neighbor negativity that peaks at criticality and, for XXZ, the next-nearest-neighbor signal as well. Side-by-side heatmaps versus independent MPS, bootstrap error bars on λ_min, and energy/correlation agreement make the central claim solid.\n\nWhat is new is the concrete integration and the full experimental map, not the individual ingredients. PPT, QOT, and brick-wall preparation are all known; the value is showing they work together on present-day superconducting hardware for mixed-state pairwise entanglement near quantum critical points, including the distinction between quantum and classical long-range correlations. That is useful both for condensed-matter simulation and as a hardware benchmark.\n\nThe soft spot the stress-test flags is real but proportionate. Unmitigated data lose the weaker next-NN signal; recovery after ZNE is load-bearing, and the authors themselves note occasional unstable extrapolations. Residual bias plus sub-unity ansatz fidelity could in principle flip a near-zero eigenvalue. They mitigate this by translational averaging and by showing the mitigated results track MPS closely, including the expected criticality peak. It is a standard NISQ caveat, not a circularity or a load-bearing math error. No public code/data is a minor practical annoyance.\n\nThis is for people who care about NISQ quantum simulation of spin chains or entanglement witnesses that survive noise and mixed states. It deserves a serious referee. I would accept it with the usual request for code release and a short discussion of residual-bias bounds on the weaker λ_min signals. Worth engaging.","headline":"Solid hardware demo of PPT + overlapping tomography that fully maps two-spin entanglement across TFIM and XXZ critical points up to 20 qubits, with quantitative MPS agreement after mitigation.","tokens_in":23147,"tokens_out":517,"would_cite":true,"duration_ms":6640,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"grok-4.5","headline":"The PPT criterion plus overlapping tomography maps all two-spin entanglement at quantum criticality on noisy hardware up to 20 qubits.","keywords":["quantum phase transitions","entanglement witness","positive partial transpose","overlapping tomography","transverse-field Ising model","XXZ model","noisy intermediate-scale quantum devices","variational quantum circuits"],"falsifier":"Prepare the same critical TFIM or XXZ states on hardware or a high-fidelity simulator, reconstruct all two-spin matrices with the same tomography protocol, and check whether the mitigated λ_min values remain negative and quantitatively match independent high-accuracy MPS or exact-diagonalization benchmarks within bootstrap error bars; a systematic sign flip or large quantitative mismatch would falsify the claim.","tokens_in":23131,"feed_emoji":"⚛️","tokens_out":725,"duration_ms":7887,"temperature":0.7,"pith_summary":"Quantum phase transitions produce highly entangled ground states, but standard entanglement entropy is hard to interpret for mixed or noisy states and costly to measure. This paper shows that the Positive Partial Transpose (PPT) test, applied to every two-spin reduced density matrix reconstructed by overlapping state tomography, is a practical and scalable witness of pairwise entanglement that works for pure and mixed states alike. The authors prepare critical ground states of the transverse-field Ising and XXZ chains with variational brick-wall circuits of up to 20 qubits on a superconducting processor, apply readout and zero-noise error mitigation, and recover the full spatial map of two-spin entanglement. Negative eigenvalues of the partial transpose peak near the critical points, match matrix-product-state benchmarks after mitigation, and cleanly separate quantum from classical correlations. The result supplies a model-independent, near-term-compatible tool both for diagnosing entanglement structure in condensed-matter simulations and for benchmarking quantum hardware.","feed_headline":"PPT maps two-spin entanglement at criticality on 20 qubits","feed_subtitle":"Overlapping tomography plus error mitigation recovers critical pairwise entanglement matching exact benchmarks","key_machinery":"Positive Partial Transpose (PPT) criterion: for any two-spin reduced density matrix ρ_AB, form the partial transpose with respect to one spin; a negative eigenvalue λ_min < 0 is necessary and sufficient for entanglement of two qubits and supplies the negativity |λ_min| as a quantitative witness, obtained for all pairs via O(log N) overlapping tomography measurements.","core_discovery":"The PPT criterion combined with quantum overlapping tomography efficiently reconstructs every two-spin reduced density matrix of a many-body state prepared on noisy hardware and certifies pairwise entanglement whenever the smallest eigenvalue of the partial transpose is negative. Applied to variationally prepared critical states of the TFIM and XXZ models (N≤20), the witness yields statistically significant negative eigenvalues that are strongest nearest-neighbor (and next-nearest for XXZ), peak at the quantum phase transitions, and agree with exact MPS results after error mitigation.","pith_inferences":[],"forward_implications":[],"fun_headline_variants":["PPT criterion maps two-spin entanglement peaking at criticality on 20 qubits","Overlapping tomography certifies pairwise entanglement in critical TFIM and XXZ states","Negative PPT eigenvalues detect nearest-neighbor entanglement at quantum phase transitions","Error-mitigated PPT recovers two-spin entanglement matching MPS for N≤20 critical states","Variational circuits on quantum hardware witness two-spin entanglement across QPTs via PPT"],"cache_read_input_tokens":16512,"weakest_assumption_plain":"That the limited-depth variational circuits plus the chosen readout and zero-noise extrapolation steps recover the true ground-state two-spin matrices near criticality closely enough that residual noise does not flip the sign of the smallest partial-transpose eigenvalue.","fun_headline_variants_meta":{"raw":{"variants":["PPT criterion maps two-spin entanglement peaking at criticality on 20 qubits","Overlapping tomography certifies pairwise entanglement in critical TFIM and XXZ states","Negative PPT eigenvalues detect nearest-neighbor entanglement at quantum phase transitions","Error-mitigated PPT recovers two-spin entanglement matching MPS for N≤20 critical states","Variational circuits on quantum hardware witness two-spin entanglement across QPTs via PPT"]},"model":"grok-4.5","effort":"low","cost_usd":0.004906,"raw_usage":{"total_tokens":1355,"prompt_tokens":710,"num_sources_used":0,"completion_tokens":105,"cost_in_usd_ticks":49060000,"prompt_tokens_details":{"text_tokens":710,"audio_tokens":0,"image_tokens":0,"cached_tokens":256},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":540,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":710,"tokens_out":105,"duration_ms":7179,"temperature":1.0,"reasoning_tokens":540,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-07-13T05:25:40.776337+00:00","model_set":{"reader":"grok-4.5"},"falsifier":"Prepare the same critical TFIM or XXZ states on hardware or a high-fidelity simulator, reconstruct all two-spin matrices with the same tomography protocol, and check whether the mitigated λ_min values remain negative and quantitatively match independent high-accuracy MPS or exact-diagonalization benchmarks within bootstrap error bars; a systematic sign flip or large quantitative mismatch would falsify the claim.","supporting_citations":[],"review_version":1}