{"id":"6fb2add1-86ce-4cf5-8840-693a1d771a26","arxiv_id":"2606.14204","paper_version":2,"verdict":"UNVERDICTED","confidence":"LOW","novelty_score":6.0,"correctness_risk":"unknown","formal_verification":"none","parameter_count":0,"one_line_summary":"A qubit-reuse protocol estimates partial-transpose moments p_2 to p_K simultaneously to additive error ε using O(K log K / ε²) copies and at most 2n+1 active qubits independent of K, with matching Ω(K/ε²) lower bounds.","lead":"The paper gives a protocol to estimate many partial-transpose moments of a quantum state at once while using a fixed number of active qubits that does not grow with the number of moments. A smart generalist might read it because the result addresses practical limits on working memory when characterizing entanglement on near-term quantum hardware.","discovery_kind":"new_method","skeptic_critique":{"model":"grok-4.3","headline":"The sequential qubit-reuse decomposition of the partial-transpose permutation must be shown to remain within 2n+1 active qubits for arbitrary K without introducing K-dependent errors.","rationale":"The reader's weakest assumption correctly isolates the single technical step whose failure would invalidate the headline memory-independence result. Because the initial review had only the abstract, the concern could not be resolved; the concrete test above directly checks whether that step holds in the full manuscript.","tokens_in":1808,"tokens_out":351,"duration_ms":15875,"concrete_test":"Extract the explicit qubit-reuse circuit (or inductive construction) from the methods/appendix; simulate its active-qubit count and fidelity for K=4 and K=32 on a 2-qubit NPT state; confirm that peak simultaneous active qubits stays ≤5 and that the estimated moments remain within additive ε of the exact values with the stated copy budget.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The central claim rests on a sequential realization of the partial-transpose permutation (abstract) that re-uses qubits so that at most 2n+1 are ever simultaneously active, independent of moment order K. This decomposition is asserted to exist and to support the O(K log K / ε²) copy bound while preserving the required measurement statistics. If the construction either (a) requires additional active qubits once K exceeds some threshold or (b) accumulates uncontrolled coherent or incoherent errors that grow with K, then both the memory-independence statement and the uniform-error guarantee fail. The provided abstract states the bound but supplies no explicit circuit or inductive argument; the full manuscript must therefore be checked for the precise sequence of local gates and reuse steps.","agreement_with_reader":"agree"},"referee_report":{"model":"grok-4.3","summary":"The manuscript studies simultaneous estimation of partial-transpose moments p_j(ρ_AB) = Tr[(ρ_AB^{T_B})^j] for j = 2 to K of an unknown n-qubit bipartite state. It claims a sequential qubit-reuse realization of the partial-transpose permutation using at most 2n+1 active qubits independent of K, achieving uniform additive error ε with total copy complexity O(K log K / ε²). It also proves matching lower bounds Ω(K/ε²) in the worst case and on an explicit isospectral two-qubit NPT family.","tokens_in":1988,"tokens_out":396,"duration_ms":10762,"significance":"If the central construction and bounds hold, the result characterizes the copy complexity of the partial-transpose moment hierarchy up to logarithmic factors and extends simultaneous nonlinear estimation techniques from ordinary state powers to partial-transpose spectral data under an active-memory constraint independent of moment order. The explicit NPT family lower bound is a concrete strength.","major_comments":[{"comment":"The central claim of memory independence rests on the existence of a sequential qubit-reuse decomposition of the partial-transpose permutation that never exceeds 2n+1 simultaneously active qubits for arbitrary K (abstract and § on the realization). The manuscript must supply the explicit inductive construction, gate sequence, or reuse schedule together with a proof that no K-dependent coherent or measurement errors are introduced; without this, both the O(K log K / ε²) upper bound and the uniform-error guarantee cannot be verified.","section":"abstract and qubit-reuse protocol section"}],"minor_comments":[],"recommendation":"major_revision","confidential_remarks":"The provided abstract states the memory bound and complexity but supplies no circuit diagram or inductive argument; the full derivations, error analysis, and verification of the 2n+1 qubit bound are not visible in the excerpt, which aligns with the low-confidence assessment on soundness."},"author_rebuttal":{"model":"grok-4.3","summary":"We thank the referee for their careful reading of the manuscript and for the positive assessment of its significance. We address the single major comment below.","responses":[{"response":"We agree that the explicit inductive construction, gate sequence, and reuse schedule, together with the accompanying error analysis, must be supplied in full detail to allow verification of the memory-independence claim. In the revised manuscript we will insert a new subsection (immediately following the current high-level description of the protocol) that presents the inductive construction of the sequential qubit-reuse decomposition of the partial-transpose permutation. The subsection will contain: (i) the base case for K=2, (ii) the inductive step that re-uses at most one additional ancilla while keeping the total active-qubit count ≤ 2n+1 for any K, (iii) the explicit gate sequence and measurement schedule, and (iv) a short proof that the construction introduces no K-dependent coherent or measurement errors. This addition will directly support both the stated copy-complexity upper bound and the uniform additive-error guarantee.","revision_made":"yes","referee_comment":"[abstract and qubit-reuse protocol section] The central claim of memory independence rests on the existence of a sequential qubit-reuse decomposition of the partial-transpose permutation that never exceeds 2n+1 simultaneously active qubits for arbitrary K (abstract and § on the realization). The manuscript must supply the explicit inductive construction, gate sequence, or reuse schedule together with a proof that no K-dependent coherent or measurement errors are introduced; without this, both the O(K log K / ε²) upper bound and the uniform-error guarantee cannot be verified."}],"tokens_in":1363,"tokens_out":360,"duration_ms":15761,"standing_objections":[]},"desk_editor":{"model":"grok-4.3","letter":"The main thing to know is that this work gives an upper bound on simultaneous estimation of partial-transpose moments p2 through pK using a sequential qubit-reuse scheme that keeps active qubits fixed at 2n+1 regardless of K, with total copy cost O(K log K / ε²), and shows Ω(K/ε²) copies are necessary in the worst case, including on an explicit two-qubit NPT family where ordinary moments stay constant.\n\nWhat is new is the adaptation of simultaneous nonlinear estimation to the partial-transpose permutation under an active-memory constraint that does not scale with moment order, plus the concrete NPT example that separates PT moments from ordinary ones. The paper does well by putting the algorithmic construction and the information-theoretic lower bounds in one place, which makes the scaling claim sharper than abstract-only statements.\n\nThe soft spot is the qubit-reuse decomposition itself. The abstract asserts that the partial-transpose permutation can be realized sequentially without exceeding 2n+1 active qubits or introducing K-dependent errors, but if the full paper only sketches the idea without an inductive argument or explicit gate sequence that preserves the required statistics, the memory-independence claim rests on an unshown step. The uniform additive error guarantee and the copy bound both depend on that step working cleanly; any hidden overhead or error accumulation would change the practical picture.\n\nThis is aimed at people working on moment-based entanglement detection or state characterization on hardware with tight active-qubit limits. A reader already comfortable with ordinary moment estimation would pick up the PT extension and the lower-bound construction quickly.\n\nIt deserves peer review so the protocol details and error analysis can be checked directly rather than desk-rejected on the abstract alone.","headline":"The paper claims a qubit-reuse protocol for simultaneous partial-transpose moment estimation capped at 2n+1 active qubits independent of K, plus near-matching lower bounds, but the construction needs explicit verification.","tokens_in":2483,"tokens_out":430,"would_cite":false,"duration_ms":27837,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"grok-4.3","headline":"A qubit-reuse circuit estimates all partial-transpose moments up to order K using at most 2n+1 active qubits independent of K.","keywords":["partial-transpose moments","qubit reuse","simultaneous estimation","copy complexity","negative partial transpose","quantum memory","moment hierarchy","entanglement witnesses"],"falsifier":"An explicit circuit decomposition or numerical check showing that, for some K larger than 2, realizing the K-th partial-transpose permutation on n qubits requires more than 2n+1 active qubits at once.","tokens_in":2701,"feed_emoji":"⚛️","tokens_out":804,"duration_ms":22803,"temperature":0.7,"pith_summary":"The paper develops a protocol for estimating the partial-transpose moments p_j of an unknown n-qubit bipartite state for all j from 2 to K at once. It realizes the required partial-transpose permutation sequentially so that only 2n+1 qubits need to be active at any time, no matter how large K becomes. The same protocol reaches uniform additive error ε across all these moments using a total of O(K log K / ε²) copies. The authors also prove matching lower bounds of Ω(K / ε²) copies that hold in the worst case and on a specific two-qubit NPT family. A reader would care because these moments supply entanglement information that ordinary state moments miss, and the fixed-memory feature matters for devices whose active-qubit count is strictly limited.","feed_headline":"Qubit reuse keeps memory fixed while estimating all partial-transpose moments","feed_subtitle":"Sequential protocol reaches uniform ε accuracy on moments up to order K with O(K log K / ε²) copies and at most 2n+1 active qubits.","key_machinery":"The sequential qubit-reuse realization of the partial-transpose permutation, which decomposes the operation into local gates and reuses to keep the number of simultaneously active qubits at most 2n+1 for any K.","core_discovery":"We give a sequential qubit-reuse realization of the partial-transpose permutation that uses at most 2n+1 active qubits, independent of K, and estimates all moments p₂,…,p_K to uniform additive error ε with total copy complexity O(K log K / ε²). We also prove two converse bounds. First, any uniformly accurate simultaneous estimator requires Ω(K/ε²) copies in the worst case. Second, the same scaling holds on an explicit isospectral two-qubit negative-partial-transpose (NPT) family whose ordinary moments are constant while the partial-transpose moments vary.","pith_inferences":["The reuse technique might extend to simultaneous estimation of other nonlinear spectral functionals that currently demand growing memory.","Hardware implementations could test whether the 2n+1 bound remains practical when n is moderate and K reaches dozens.","The NPT example indicates that partial-transpose moments can separate states that are indistinguishable by ordinary moment estimates alone."],"forward_implications":["All moments from order 2 to K are obtained to uniform additive error ε with total copy count O(K log K / ε²).","The active-qubit count remains capped at 2n+1 no matter how high the target moment order becomes.","Any simultaneous estimator that works uniformly over all moments still needs at least Ω(K / ε²) copies in the worst case.","The same linear-in-K lower bound applies even to an explicit two-qubit NPT family whose ordinary moments stay constant."],"fun_headline_variants":["Qubit reuse fixes active memory for all partial-transpose moments","Fixed qubit count enables simultaneous partial-transpose moment estimation","Memory stays bounded independent of moment order for partial-transpose estimates","Sequential qubit reuse achieves constant active memory across moment orders","Active memory independent of K for estimating multiple partial-transpose moments"],"cache_read_input_tokens":2112,"weakest_assumption_plain":"The partial-transpose permutation circuit admits a sequential decomposition into local operations and qubit reuses that never requires more than 2n+1 simultaneously active qubits, regardless of how large K grows.","fun_headline_variants_meta":{"raw":{"variants":["Qubit reuse fixes active memory for all partial-transpose moments","Fixed qubit count enables simultaneous partial-transpose moment estimation","Memory stays bounded independent of moment order for partial-transpose estimates","Sequential qubit reuse achieves constant active memory across moment orders","Active memory independent of K for estimating multiple partial-transpose moments"]},"model":"grok-4.3","cost_usd":0.005545,"raw_usage":{"total_tokens":2686,"prompt_tokens":721,"num_sources_used":0,"completion_tokens":80,"cost_in_usd_ticks":55449500,"prompt_tokens_details":{"text_tokens":721,"audio_tokens":0,"image_tokens":0,"cached_tokens":256},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":1885,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":721,"tokens_out":80,"duration_ms":12717,"temperature":1.0,"reasoning_tokens":1885,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-06-29T04:56:40.631218+00:00","model_set":{"reader":"grok-4.3"},"falsifier":"An explicit circuit decomposition or numerical check showing that, for some K larger than 2, realizing the K-th partial-transpose permutation on n qubits requires more than 2n+1 active qubits at once.","supporting_citations":[],"review_version":1}