{"id":"ec878815-3844-4739-a22e-d8769fa5139e","arxiv_id":"2506.06466","paper_version":1,"verdict":"REJECT","confidence":"HIGH","novelty_score":5.0,"correctness_risk":"high","formal_verification":"none","parameter_count":2,"one_line_summary":"The authors claim an LOCC protocol distinguishes any two orthogonal entangled two-qubit states sequentially with success >1/2 per round while preserving finite entanglement, but the general-case proof uses a false equality.","lead":"This paper proposes a sequential protocol in which many pairs of observers each locally measure two entangled two-qubit states, guessing which state was sent while leaving some entanglement intact. The authors claim success always beats random guessing for any number of rounds, but the proof for the general two-state case relies on a false assumption.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Appendix B's witness proof uses the false Schmidt-basis equality Tr[σ3⊗σ3ρ0_b]=-1; in the computational basis of Eq. (1) it is not -1, so the arbitrary-round witness claim behind Theorem 3 is unsupported.","rationale":"The reader's weakest-assumption analysis identifies exactly the load-bearing flaw in the general-case argument. Appendix B is the only support for the claim that arbitrary numbers of sequential pairs can witness entanglement, and its key equality Tr[σ3⊗σ3ρ0_b]=-1 is false in the computational basis of Eq. (1). The specific-case Theorems 1 and 2 appear internally consistent, and those states were chosen so that the false equality would hold; this explains why the specific-case construction works. But the paper's abstract and conclusion advertise the general result for any two orthogonal entangled two-qubit states, and that result requires Theorem 3. Since Appendix B's chain collapses and Appendix C's success-probability proof relies on numerical plots rather than a complete analytic proof, the central claim of the paper is not established as written. A future revision might replace the witness with one adapted to the actual Schmidt bases or use logarithmic negativity directly, but that is not present in this manuscript. The verdict REJECT is therefore appropriate, and my stress-test does not change it.","tokens_in":30923,"tokens_out":18298,"duration_ms":185747,"concrete_test":"Recompute the first lines of Appendix B using the actual states from Eq. (1), not their Schmidt forms. For µ1=µ2=1/2, θ=π/2, verify that Tr[σ3⊗σ3ρ0_1]=+1 and Tr[σ3⊗σ3ρ0_2]=-1, contradicting the assumed common value -1. Then propagate one round with λ1→0 through Eq. (A1) and check whether λ2 defined by Eq. (22) lies in (0,1): with the standard σ2, the denominator Tr[σ2⊗σ2ρ0_b] is negative for both states, so the witness inequality (20) cannot be satisfied by any g∈[0,1]; with the opposite sign convention, the required g for state b=1 exceeds 1. This single calculation settles that the chain in Appendix B does not establish arbitrarily many witnessing rounds.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The general-case theorem rests on Appendix B's claim that arbitrary rounds can witness entanglement. After Eq. (22), the authors infer Tr[σ3⊗σ3ρ0_b]=-1 for both initial states from the Schmidt form √m|01>+√(1-m)|10>. This conflates the Schmidt basis with the fixed computational basis used in Eq. (1). Direct computation from Eq. (1) gives Tr[σ3⊗σ3ρ0_1]=µ1-(1-µ1)cos2θ=1-2(1-µ1)cos²θ, which equals -1 only at µ1=0 or θ=0, both outside the entangled regime µ1,2∈(0,1), θ∈(0,π/2]. For ρ0_2 the value is -1+2(1-µ2)cos²θ, which is generically not -1 either. Hence the chain λ1→0 ⇒ λ2→0 ⇒ λ3→0 has no valid starting point. The failure is concrete: for the allowed example µ1=µ2=1/2, θ=π/2, the two initial states are (|00>+|11>)/√2 and (|01>-|10>)/√2, whose σ3⊗σ3 expectations are +1 and -1, not both -1. With the standard sign of σ2, Tr[σ2⊗σ2ρ0_b]<0 for both states, so the actual witness condition Tr[W_b^0ρ0_b]<0 is impossible for any g∈[0,1]; with the opposite sign convention, b=1 requires g>2>1. Thus the witness construction (19)-(22) cannot certify the initial states, and Theorem 3's general entanglement-retention claim is not proved by the manuscript.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proposes a sequential state discrimination protocol (SSDSE) in which multiple pairs of observers distinguish two orthogonal entangled two-qubit pure states by unsharp local measurements with classical communication, while aiming to keep the post-measurement states entangled at every round. For a particular two-state family (Sec. IV) the authors derive the per-round success probability and the residual logarithmic negativity; for general orthogonal entangled two-qubit states (Sec. V) they claim, in Theorem 3, that arbitrarily many rounds can succeed with probability above random guessing while entanglement is witnessed in each round. The general-case proof relies on a witness construction in Sec. V A and on Appendices B and C.","tokens_in":31371,"tokens_out":16000,"duration_ms":161401,"significance":"The specific-case results are concrete and checkable: Theorems 1 and 2 give P^k_suc = 1/2 + λ_k^2/2 and E^k_b = log2[1 + 2ϑ_b(1 − S_k)], and direct calculation confirms these formulae. If the general claim were established, the paper would be a useful contribution to the recent literature on recycling quantum correlations by sequential weak measurements. However, the general-case Theorem 3 is the central new assertion of the paper, and its proof is invalid at a load-bearing point: the witness construction does not certify the initial states in the computational basis used by the protocol, and the arbitrary-round success-probability proof is not rigorous. The general claim is therefore not established by this manuscript.","major_comments":[{"comment":"The chain argument for arbitrarily many rounds is based on the claim that Tr[σ3⊗σ3 ρ0_b] = −1 for both initial states. Appendix B obtains this by writing a pure entangled state, up to local unitary, as √m|01⟩+√(1−m)|10⟩. But the states in Eq. (1) are fixed in a computational basis that is not the Schmidt basis of either state, and the protocol's measurements are defined in that same computational basis. In the computational basis, for |Φ1⟩=√µ1|00⟩+√(1−µ1)|1ς1⟩ one obtains Tr[σ3⊗σ3 ρ0_1] = µ1−(1−µ1)cos2θ, which equals −1 only for µ1=0 or special θ, outside the entangled regime µ1∈(0,1). For |Φ2⟩=√µ2|01⟩+√(1−µ2)|1ς1⊥⟩ one obtains Tr[σ3⊗σ3 ρ0_2] = −µ2+(1−µ2)cos2θ, which is also not −1 in general. Hence the inference λ1→0 ⇒ λ2→0 ⇒ ⋯ in Eq. (B1) has no valid base, and the claimed existence of witnesses W^k_b for arbitrarily many rounds is unsupported.","section":"Appendix B; Sec. V A"},{"comment":"The witness condition is transcribed with the wrong inequality direction when Tr[σ2⊗σ2 ρ] is negative. For W^k_b = 1/4(I + σ3⊗σ3 − g^k_2 σ2⊗σ2), the condition Tr[W^k_b ρ] < 0 is 1 + Tr[σ3⊗σ3 ρ] − g Tr[σ2⊗σ2 ρ] < 0. For both initial states in Eq. (1), with the standard Pauli σ2 and in the computational basis, Tr[σ2⊗σ2 ρ0_b] < 0; for example, at µ1=µ2=1/2, θ=π/2, Tr[σ2⊗σ2 ρ0_1] = Tr[σ2⊗σ2 ρ0_2] = −1. In this case the correct inequality is g < (1+Tr[σ3⊗σ3 ρ])/Tr[σ2⊗σ2 ρ], not the > in Eq. (20), and since 1+Tr[σ3⊗σ3 ρ] ≥ 0 for any state, no g∈[0,1] can make Tr[W ρ] < 0. Thus the witness operator of Eq. (19) cannot certify the entanglement of the initial states in the protocol's measurement basis, and the recursive construction in Eq. (22) has no valid starting point.","section":"Sec. V A, Eqs. (19)–(22)"},{"comment":"The proof that Q^k_b > 0 for θ∈(π/4,π/2] and hence that P^k_suc > 1/2 for all k is not a proof as written. Several key steps are justified only by numerical inspection for k = 2 (Figs. 2–5) or by assertions such as 'it can be shown' and 'it can be checked numerically' for functions h, a, c whose sign properties are stated to hold 'in the limit λ_k→0' without closed-form demonstration. The induction to arbitrary k also requires the sequence {λ_k} to be strictly increasing while satisfying the witness-prescribed recursion (22) and remaining in (0,1], but no argument establishes that such a sequence exists for every finite k. Consequently, Theorem 3's claim of arbitrarily many rounds with per-round success probability strictly above 1/2 is not established.","section":"Appendix C; Theorem 3"}],"minor_comments":[{"comment":"There are several typographical errors: the Introduction ends with 'lastly conclude in Sec. V B', which should refer to the conclusion section; Sec. IV says 'discriminating |κ1⟩ and |κ1⟩' where the second state should be |κ2⟩; and the sentence 'the states ¯ρb are pure entangled states of the form ¯ρb = |κb⟩⟨κb|' uses an undefined index b before the states are introduced.","section":"Sec. I and Sec. IV"},{"comment":"The parameterization in Eq. (1) is taken from Ref. [91] and the entangled regime is stated as µ1,2∈(0,1), θ∈(0,π/2]. The paper should explicitly note that this covers all orthogonal entangled pairs only up to local unitary, since the subsequent witness construction is basis-dependent; the present wording could mislead the reader into thinking the computational basis is the Schmidt basis.","section":"Sec. II B 1, Eq. (1)"},{"comment":"The witness form in Eq. (18) is cited from Refs. [109,118,119] with |g_q|≤1. The paper then restricts to g_1=0, g_3=1, g_2∈[0,1] without justifying that this restricted family can witness every entangled two-qubit state in the computational basis; as shown in the major comments, it cannot for the states considered.","section":"Sec. V A, Eq. (18)"},{"comment":"The definition of unsharp measurement in Eq. (5) requires λ∈(0,1), but later the paper also discusses λ=1 and λ→0. Please clarify the allowed range of λ in each statement and whether λ=0 corresponds to a valid measurement or only to a limiting case.","section":"Sec. III, Eq. (5)"}],"recommendation":"reject","confidential_remarks":"The specific-case theorems in Sec. IV are sound and could be part of a publishable note, but the manuscript's headline claim is the general Theorem 3, and its proof fails at the witness construction and at the numerical/non-rigorous induction. Because the central claim is not established and the required repair would involve reworking the witness and the entire arbitrary-round argument, I recommend rejection rather than minor or major revision. I see no issue with citation practice or overlap beyond the authors' own previous recycling-of-entanglement line of work."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The specific-case part of this paper is genuinely good. Theorems 1 and 2 for the two chosen states are correct; I checked the success probability formula and the logarithmic negativity expression, and they hold. The idea of using unsharp measurements to trade a bit of discrimination success for entanglement retention is a sensible extension of the recycling literature, and the presentation is honest about the trade-off.\n\nThe problem is the general case. Theorem 3 is the headline, and it is not established. Appendix B argues that Tr[σ3⊗σ3ρ0_b] = −1 for both initial states by invoking the Schmidt decomposition. That equality only holds in the Schmidt basis, not in the computational basis used for the states in Eq. (1). For the allowed example µ1=µ2=1/2, θ=π/2, the two initial states are (|00⟩+|11⟩)/√2 and (|01⟩−|10⟩)/√2, for which the σ3⊗σ3 expectations are +1 and −1, not both −1. The stress-test note is right.\n\nThis is not a minor technical slip. The chain λ1→0 ⇒ λ2→0 ⇒ ⋯ ⇒ λk→0 that guarantees witness existence for arbitrarily many rounds depends on that equality. Without it, there is no proof that the sharpness parameters stay within [0,1], and the whole arbitrary-round claim collapses. There is also a sign issue: Tr[σ2⊗σ2ρ0_b] is negative over the full parameter range of Eq. (1), so the inequality (20) has the wrong direction. The proposed witness cannot certify the initial states at all for generic parameters. Appendix C's numerical checks without code would also need tightening in any revision, but the main obstruction is Appendix B.\n\nIn short: a correct and publishable special-case result, wrapped in an unsupported general claim. A reader working on sequential state discrimination or entanglement recycling might find the special case useful, but the paper's stated generality is not proven.\n\nRecommendation: send it to peer review. The specific-case result deserves an outlet, and the general-case flaw is concrete enough that a referee can point to it precisely. The authors should either fix the witness construction, restrict Theorem 3 to cases where the computation actually works, or drop the general claim.","headline":"Correct special-case result, but Theorem 3 rests on a basis-dependent error in Appendix B that breaks the arbitrary-round claim.","tokens_in":31852,"tokens_out":9666,"would_cite":false,"duration_ms":79203,"reading_group":"maybe","serious_thinker":"no","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"For any two entangled two-qubit states, sequential observers can beat random guessing indefinitely while preserving entanglement.","keywords":["sequential state discrimination","local operations and classical communication","minimum-error state discrimination","unsharp measurement","entanglement preservation","entanglement witness","logarithmic negativity","two-qubit pure states"],"falsifier":"Take the paper's own initial state |Φ1⟩ = √µ1|00⟩ + √(1−µ1)|1⟩(cosθ|0⟩ + sinθ|1⟩) and compute Tr[σ3⊗σ3|Φ1⟩⟨Φ1|] = µ1 − (1−µ1)cos2θ. For µ1 = 0.4 and θ = π/3 this equals 0.7, not −1, so the Appendix B basis for concluding λ2→0 fails for that generic choice; running the witness inequality (20) explicitly for round 2 with these parameters would settle whether arbitrarily many rounds can all certify entanglement for every state in the claimed family.","tokens_in":30743,"feed_emoji":"♾️","tokens_out":6361,"duration_ms":62493,"temperature":0.7,"pith_summary":"This paper asks whether sequential state discrimination—multiple pairs of observers measuring the same bipartite state one after another—can be done without destroying the entanglement being measured. Standard protocols can distinguish any two orthogonal pure states perfectly, but the sharp measurement collapses the state into a separable one. The authors replace sharp measurements with unsharp (noisy) ones and show that in each round the success probability stays above the 1/2 random-guessing level while the post-measurement states keep finite entanglement. For a special family of two-qubit states they compute the residual entanglement exactly and can push the success probability arbitrarily close to 1. For arbitrary orthogonal entangled two-qubit states they claim a witness-operator argument certifies entanglement retention for any number of rounds.","feed_headline":"Entanglement survives unlimited rounds of quantum state guessing","feed_subtitle":"Sequential pairs can identify two entangled qubits better than chance at every step without draining their entanglement.","key_machinery":"The central mechanism is the unsharp POVM: each sharp rank-one projector P_j is replaced by Q_j = λP_j + (1−λ) I/d, so part of the measurement acts as a ‘do nothing’ channel. Tuning λk controls the trade-off between information gain and entanglement damage. The state is propagated between rounds by the von Neumann–Lüders update ρ_k^b = Σ_j √(O_j^k) ρ_{k−1}^b √(O_j^k). Entanglement retention is quantified by logarithmic negativity for the special family and certified by the witness operator W_k^b for the general family; the next round’s sharpness parameter is set to λ_{k+1} = (1+ε_k) times the maximum of the ratios (1+Tr[σ3⊗σ3 ρ_k^b])/Tr[σ2⊗σ2 ρ_k^b], designed to keep the witness negative.","core_discovery":"The central claim is that for any two orthogonal, entangled, two-qubit pure states prepared with equal probability, there exist unsharp one-way LOCC measurements such that an arbitrary number of sequential pairs can discriminate the states with average success probability strictly greater than 1/2 at every round, while the state handed on by each pair remains entangled. The protocol starts from the optimal sharp LOCC measurement for two orthogonal two-qubit states and replaces each projector with a noisy version whose sharpness parameter λk is chosen round by round. For the special family |κ1⟩ and |κ2⟩, the success probability at round k is 1/2 + λ̄_k²/2 and the logarithmic negativity of each post-measurement state is log2[1 + 2ϑ_b(1 − S_k)], both strictly positive for λ̄_k ∈ (0,1). For the general family, entanglement is certified by a witness operator W_k^b = (1/4)(I⊗I + σ3⊗σ3 − g2^k σ2⊗σ2), with g2^k chosen so that Tr[W_k^b ρ_k^b] < 0, and the success probability is claimed to exceed 1/2 for arbitrarily many rounds.","pith_inferences":["If the missing σ3⊗σ3 condition is enforced by rotating the initial states into a basis where each is Schmidt anti-correlated, the witness construction might extend cleanly to all entangled pairs; this is an editorial projection, not a claim in the paper.","One testable extension is to apply the same unsharp-measurement replacement to unambiguous sequential discrimination or to multipartite states; the paper only treats the minimum-error two-qubit case.","The numerical checks at k = 2 in the general-case proof could likely be replaced by a fully analytic computation of the R and R′ recursion coefficients in closed form, which would remove the remaining numerical step.","The sharpness parameter acts like an entanglement budget that each observer can choose independently; for the special family the budget is not depleted by previous rounds, suggesting a modular resource interpretation that the paper does not develop."],"forward_implications":["If the main theorem is correct, arbitrarily many receiver pairs can extract classical information from an entangled two-qubit resource while preserving a usable entangled state for later tasks.","For the special family, each round's success probability depends only on that round's sharpness parameter, so earlier observers' choices do not degrade later discrimination power.","A final pair that declines to decode can still use the remaining entangled state, since every two-qubit entangled state is distillable.","The protocol works under one-way LOCC, so it uses the same experimental resources as standard sequential state discrimination.","For the special family, the success probability can be made arbitrarily close to 1 by taking λ̄_k close to 1 while keeping nonzero logarithmic negativity."],"supporting_citations":[{"why":"Supplies the parametrization of two orthogonal two-qubit states and the optimal sharp LOCC measurement that the protocol unsharpens.","marker":"[91]"},{"why":"Provides the von Neumann–Lüders update rule used to propagate post-measurement states between rounds.","marker":"[117]"},{"why":"Defines logarithmic negativity, used to quantify the residual entanglement in the special-family analysis.","marker":"[114]"},{"why":"Gives the entanglement-witness form used to certify entanglement in the general case.","marker":"[109]"},{"why":"Supplies the witness-operator criterion with Pauli-correlation terms on which the general-case witness is built.","marker":"[118]"},{"why":"States that every two-qubit entangled state is distillable, motivating the goal of preserving entanglement.","marker":"[104]"},{"why":"Introduces sequential state discrimination by multiple observers, the task this paper extends to entanglement preservation.","marker":"[87]"}],"fun_headline_variants":["Unlimited rounds of qubit discrimination sustain entanglement","Sequential pairs identify qubits while keeping entanglement alive","Endless state guessing with entanglement intact, better than chance","Arbitrary rounds of local measurements preserve qubit entanglement","Entanglement survives each step of unbounded sequential guessing"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that both initial states have Tr[σ3⊗σ3ρ0_b] = −1; in the paper's own parametrization of general two-qubit states this equality holds only for special parameter choices, so the chain λ2→0, λ3→0, ... that keeps the witness valid for arbitrarily many rounds is not guaranteed for all states the theorem claims to cover.","fun_headline_variants_meta":{"raw":{"variants":["Unlimited rounds of qubit discrimination sustain entanglement","Sequential pairs identify qubits while keeping entanglement alive","Endless state guessing with entanglement intact, better than chance","Arbitrary rounds of local measurements preserve qubit entanglement","Entanglement survives each step of unbounded sequential guessing"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000349,"raw_usage":{"total_tokens":1902,"prompt_tokens":938,"completion_tokens":964,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":554,"completion_tokens_details":{"reasoning_tokens":897}},"tokens_in":554,"tokens_out":964,"duration_ms":10104,"temperature":1.0,"reasoning_tokens":897,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T05:57:11.048433+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take the paper's own initial state |Φ1⟩ = √µ1|00⟩ + √(1−µ1)|1⟩(cosθ|0⟩ + sinθ|1⟩) and compute Tr[σ3⊗σ3|Φ1⟩⟨Φ1|] = µ1 − (1−µ1)cos2θ. For µ1 = 0.4 and θ = π/3 this equals 0.7, not −1, so the Appendix B basis for concluding λ2→0 fails for that generic choice; running the witness inequality (20) explicitly for round 2 with these parameters would settle whether arbitrarily many rounds can all certify entanglement for every state in the claimed family.","supporting_citations":[{"cited_title":"Experimental multiparty sequential state discrimina- tion,","cited_arxiv_id":null,"evidence_quote":"Supplies the parametrization of two orthogonal two-qubit states and the optimal sharp LOCC measurement that the protocol unsharpens."},{"cited_title":"Logarithmic negativity: A full entangle- ment monotone that is not convex,","cited_arxiv_id":null,"evidence_quote":"Provides the von Neumann–Lüders update rule used to propagate post-measurement states between rounds."},{"cited_title":"Lo- cal entanglement transfer to multiple pairs of spatially separated observers,","cited_arxiv_id":null,"evidence_quote":"Defines logarithmic negativity, used to quantify the residual entanglement in the special-family analysis."},{"cited_title":"Arbitrarily many inde- pendent observers can share the nonlocality of a single maximally entangled qubit pair,","cited_arxiv_id":null,"evidence_quote":"Gives the entanglement-witness form used to certify entanglement in the general case."},{"cited_title":"Teleporting an un- known quantum state via dual classical and Einstein- Podolsky-Rosen channels,","cited_arxiv_id":null,"evidence_quote":"States that every two-qubit entangled state is distillable, motivating the goal of preserving entanglement."},{"cited_title":"Quantum receiver for phase-shift keying at the single-photon level,","cited_arxiv_id":null,"evidence_quote":"Introduces sequential state discrimination by multiple observers, the task this paper extends to entanglement preservation."}],"review_version":1}