{"id":"e4777bdb-c599-4c66-a2f9-b5341a593d46","arxiv_id":"2411.14537","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"An experiment with light demonstrates the optimal fixed-rate-of-inconclusive-outcomes (FRIO) discrimination for N=2,3,5,7 symmetric qubit states using state separation followed by minimum-error measurement.","lead":"This paper reports an optical experiment that demonstrates the optimal strategy for telling apart N symmetric quantum states when a fixed fraction of measurements may give no answer. The method combines a probabilistic step that makes the states easier to distinguish with a minimum-error measurement, and the measured error rates match the theoretical optimum.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The benchmark for optimality is the ideal pure-state curve, but the produced states are depolarized; the true FRIO optimum for the measured mixed states is not computed, so the claim of reaching the minimum error rate is not yet established.","rationale":"The reader's weakest_assumption identifies exactly the same load-bearing concern: the theoretical benchmark assumes ideal symmetric pure states, while the actual states are depolarized and mixed. This is the right concern because the claim is an experimental demonstration of optimal FRIO discrimination. If the comparison benchmark is the ideal curve rather than the true optimum for the realized states, then strong agreement with the curve does not logically imply that the implementation attains the minimum error rate for those states. The missing piece is a benchmark computed from the characterized density matrices, which the test above would provide. The paper otherwise has independent support: the theoretical derivation of Eq. (13) is standard and the two-step decomposition is clearly explained; the experimental characterization of the separated states is thorough. Missing error bars in Fig. 5 are a secondary issue, but the standard deviations are stated to be ~10^-3, so the main weakness remains the benchmark mismatch. The conditional verdict remains appropriate: the paper is valuable but should not be fully accepted until the comparison against the true optimum for the actual states is performed.","tokens_in":18142,"tokens_out":9008,"duration_ms":90330,"concrete_test":"Compute the FRIO-optimal error curve for the experimentally characterized states. Using the measured V_j and ϕ_j from Section V A (via Eq. (18)) for each N and θ'_t, solve the semidefinite program that minimizes Pe subject to a fixed inconclusive rate Q for the set {ρ_j} (or use the known FRIO solution for mixed symmetric states if available), and compare the reported [Pe]_expt against this true optimum at each Q. In particular, check the N=2, Q=Q_MC endpoint: if the measured error is nonzero, verify that it equals the minimal positive error for the two mixed states. If the data lie within uncertainty of the true mixed-state optimum, the claim is supported; if they are close to the ideal pure-state curve but above the true optimum, the claim of reaching the minimum error rate is not supported.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim, that the implementation closely reaches the minimum error rate for fixed Q, is tested against Eq. (13), the FRIO optimum for ideal symmetric pure states |β_j(θ')>. However, the successfully separated states are not exactly these pure states. Section V A characterizes them as ρ_j(θ'_t) in Eq. (18) with measured visibilities V_j and phases ϕ_j. Depolarization of the transmissive LCD (Eq. B2) makes V_j < sin 2θ'_t for the larger gray levels, so the actual states are mixed and less distinguishable than the ideal ones. The ME measurement applied in the second step is the optimal POVM for the ideal pure states, not necessarily for the actual mixed states. Consequently, agreement with the ideal curve does not establish that the achieved error rate is the minimum possible for the states actually produced. The problem is most acute at the endpoint Q = Q_MC: for N=2 the ideal curve gives P_min^e = 0, but unambiguous discrimination of two overlapping mixed states is impossible, so the true minimum for the actual states is strictly positive. The paper does not compute this true optimum, leaving open the possibility that the data lie above it, i.e., that the implementation is suboptimal for the actual states. This is a benchmark mismatch, not an internal inconsistency, but it directly affects the paper's headline claim.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper presents an experimental demonstration of optimal fixed-rate-of-inconclusive-outcomes (FRIO) discrimination for N=2, 3, 5, and 7 equally likely symmetric qubit states encoded in photonic path modes. The implementation follows a two-step strategy: first, a probabilistic optimal state separation stage based on a programmable spatial light modulator, which fixes the inconclusive rate Q; second, a minimum-error (ME) measurement on the successfully separated states, implemented via an optical Fourier transform and detector arrays. The theoretical part derives the error probability Pe(θ') and inconclusive rate Q(θ') as functions of the separation angle and obtains the FRIO benchmark P_min^e(Q) of Eq. (13), interpolating between minimum-error (Q=0) and maximum-confidence/unambiguous (Q=Q_MC) discrimination. The authors compare experimental success probabilities, correct-discrimination probabilities, and the resulting error-versus-Q tradeoff with theoretical curves and conclude that the scheme closely reaches the minimum error rate for fixed Q.","tokens_in":18324,"tokens_out":3861,"duration_ms":34592,"significance":"If fully established, the paper would be a valuable addition to the experimental quantum state discrimination literature: it goes beyond the previously demonstrated two-state FRIO case to N=2,3,5,7 symmetric qubit states, and it explicitly demonstrates the two-step decomposition (probabilistic state separation followed by a minimum-error measurement) as a practical route to optimal FRIO discrimination. The theoretical formulas are standard and correctly cited, and the paper provides a careful characterization of the separated states (Eq. (18), Fig. 3), an explicit depolarization model for the SLM (Eq. (B2)), and a plausible path toward higher-dimensional extensions. The experimental data in Figs. 4 and 5 show reasonable agreement with the theoretical curves, with the caveat discussed below that the benchmark curve assumes ideal pure output states. The paper is clearly written and the experimental methodology is reproducible in its main elements.","major_comments":[{"comment":"The headline claim that the implementation 'closely reaches the minimum error rate for a fixed value of Q' is tested against Eq. (13), which is the FRIO optimum for ideal symmetric pure states |β_j(θ')>. However, the experimentally separated states are the mixed states ρ_j(θ'_t) of Eq. (18), with measured visibilities V_j that are in some cases significantly below sin 2θ'_t (notably for θ'_6 and θ'_7, as acknowledged in Section V A). The ME POVM of Eq. (9) is optimal for the ideal pure states, not necessarily for these mixed states, and the paper does not compute the true FRIO-optimal error rate for the measured states. Therefore, agreement with Eq. (13) does not by itself establish that the achieved error rate is the minimum possible for the states actually produced. This is a fixable but load-bearing gap: the authors already have the measured V_j and φ_j, so they could compute the true optimum (or a rigorous bound) for the measured mixed states and compare their data against it. Without that comparison, the central optimality claim is not fully supported.","section":"Section V A, Eq. (18), Fig. 5"},{"comment":"The benchmark mismatch is most acute at the endpoint Q = Q_MC = 0.7771. For N=2, Eq. (13) gives P_min^e(Q_MC)=0, corresponding to optimal unambiguous discrimination. But unambiguous discrimination of two nonorthogonal mixed states is impossible unless the states are orthogonal, and the measured N=2 states at θ'_7 have visibilities below the ideal value. The data point near Q=0.7771 must therefore lie strictly above the true minimum error for the actual mixed states if V_j < sin 2θ'_7. The paper does not quantify this gap; the authors should compute the true mixed-state optimum for the measured visibilities and phases, or provide a quantitative argument that the deviations are negligible within their reported error bars.","section":"Section V, Fig. 5, endpoint Q = Q_MC"},{"comment":"The experimental FRIO probabilities are constructed as [Pe]_expt = [ps]_expt (1 - [pβ_c]_expt) and Q_expt = 1 - [ps]_expt. This is consistent with Eq. (11) only if the two averages can be multiplied without introducing bias. Since [ps]_expt and [pβ_c]_expt are averaged separately over the input states and the measured quantities come from the same intensity distributions, the paper should explicitly justify that no conditioning or post-selection bias enters this product. In particular, the authors should clarify that the same set of input states and the same detected events underlie both averages, and that the camera detection efficiency compensation described in Section V B does not distort the joint statistics.","section":"Section V B and Eqs. (11)"},{"comment":"The derivation leading to Eq. (13) via Eqs. (6), (10), and (11) is algebraically clean, but the paper does not prove that the two-step POVM of Section III D is globally optimal for the FRIO problem; it relies for that on prior results (Refs. [20,21,34]). This is acceptable for an experimental demonstration, but the claim in Section VI that the paper demonstrates 'both theoretically and experimentally' the optimal FRIO discrimination is slightly overstated. The theoretical optimality is imported from the literature; the experimental contribution is the implementation and test of the two-step scheme against that known benchmark.","section":"Section III D and Eq. (13)"}],"minor_comments":[{"comment":"Please define the ordering of the ancilla basis used in the 2×2 block of Eq. (7), and state explicitly that ξ(θ') = tan θ cot θ' lies in [0,1] for θ ≤ θ' ≤ π/4, which is used later.","section":"Eq. (7)"},{"comment":"The sign convention and branch choice in the Fourier transform leading to x_k = -λf m_k/(NΔ) would benefit from an explicit statement connecting Eq. (16) to the projective measurement of Eq. (15); currently the reader must reconstruct the signs from context.","section":"Section IV C and Eq. (17)"},{"comment":"In Fig. 3, the experimental markers for θ'_6 and θ'_7 lie noticeably inside the ideal parallels. Please provide the numerical values of the measured visibilities V_j (or at least their range) either in the caption or in a small table, so the purity loss can be assessed quantitatively without reading the figure by eye.","section":"Table I and Fig. 3"},{"comment":"The attribution of V_j > sin 2θ'_t to 'inaccuracies in the preparation of the input states' is plausible but is not supported by a quantitative model; a sentence explaining which preparation imperfection (e.g., imperfect phase or amplitude in the grating masks) produces this effect would be helpful.","section":"Section V A, footnote 7"},{"comment":"The depolarizing model of Eq. (B2) describes the ancilla polarization, but the relation between the ancilla depolarization and the qubit visibility V_j in Eq. (18) is not derived explicitly. A short derivation or a reference to one would strengthen the connection between the SLM characterization and the measured states.","section":"Appendix B, Eq. (B2)"},{"comment":"There are several small typographical issues: 'th second column' in the Table I caption; 'a useful platform' in the introduction; the word 'minium' in the arXiv metadata. Also, the acronym FRIO is used in the abstract without being spelled out there; please define it at first use in the abstract or tolerate the fact that the introduction defines it.","section":"Abstract and text"}],"recommendation":"major_revision","confidential_remarks":"The paper is a solid experimental demonstration with a correct theoretical framework. The central unresolved issue is the benchmark mismatch: the experimental states are characterized as mixed (Eq. (18)), but the optimality claim is tested against the ideal pure-state curve Eq. (13). Because the authors already measure V_j and φ_j for each data point, computing the true FRIO optimum for the actual mixed states (or a rigorous bound) is a feasible and appropriate revision. If the true optimum is close to Eq. (13), the claim can be restored quantitatively; if not, the claim will need to be weakened. The N=2 endpoint at Q_MC is the clearest stress point because the ideal curve predicts zero error, which is unattainable for overlapping mixed states."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick take: this is a solid experimental demonstration of the two-step FRIO strategy for N symmetric qubit states, and the central caveat is that they benchmark against the ideal pure-state curve rather than the true optimum for their actual mixed states.\n\nWhat's new: the theory is all known—Eq. (13) is Bagan/Herzog, the separation is Chefles-Barnett—but the experiment for N=3,5,7 path-encoded qubits is new, and it's done carefully. They characterize the separated states via interference visibility and phase, show the measured success probabilities track Eq. (6), and the error-rate curves in Fig. 5 come close to the ideal prediction for all N. The two-step decomposition is clean and the optical implementation is convincing. The paper is honest about the classical-laser approximation and the depolarization of the transmissive LCD.\n\nSoft spots: the main one is exactly what the stress test says. The comparison in Fig. 5 is against P_min^e(Q) for ideal pure symmetric states. The actual states have measured visibilities below sin 2θ'_t at the larger gray levels, making them mixed. The ME POVM applied in the second step is the optimal one for the ideal states, not necessarily for the real ones, and the true FRIO optimum for the characterized ρ_j(θ'_t) is not computed. So 'closely reaches the minimum error rate' is not strictly demonstrated—agreement with the ideal curve doesn't rule out being above the true optimum. The endpoint Q=Q_MC for N=2 is the clearest case: ideal curve predicts zero error, but unambiguous discrimination of overlapping mixed states is impossible, so the true minimum is strictly positive. That said, the paper does say 'closely reaches' rather than 'achieves', and the observed deviations are small for most points. It's a benchmark mismatch, not an internal inconsistency, and it weakens the headline claim without destroying the main message.\n\nMinor: no error bars in Fig. 4 and 5, though they state the standard deviations are ~10^-3; that's acceptable but a bit lazy. The phase compensation and detector-efficiency corrections are described reasonably.\n\nBottom line: the paper is a legitimate experimental contribution. The referee should ask for a discussion of the benchmark issue—ideally compute the true optimum for the measured density matrices, or at least bound how far the actual states are from the ideal ones. That's a revision, not a rejection.\n\nI'd send it to peer review. It's useful for the quantum communication and state-discrimination community.","headline":"Solid experimental demonstration of the two-step FRIO strategy for N symmetric qubit states, with the main caveat being that optimality is benchmarked against ideal pure states rather than the actual mixed ones.","tokens_in":18991,"tokens_out":1971,"would_cite":true,"duration_ms":18439,"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":"An optical experiment demonstrates the quantum-optimal trade-off between error and inconclusive outcomes when discriminating N=2, 3, 5, and 7 symmetric qubit states.","keywords":["quantum state discrimination","fixed rate of inconclusive outcomes","state separation","minimum-error measurement","symmetric qubit states","spatial light modulator","path-encoded photonic qubits","maximum-confidence discrimination"],"falsifier":"Compute the true minimum error for the actual mixed states $\\hat\\rho_j(\\theta')$ of Eq. (18) using the measured visibilities $V_j$ and phases $\\phi_j$, and check the data at the largest separation angles ($\\theta'_6$ and $\\theta'_7$), where depolarization is strongest: if the experimental error rates lie above that recomputed optimum by more than the experimental uncertainty, the claim that the implementation closely reaches the minimum error rate for fixed $Q$ is falsified.","tokens_in":17858,"feed_emoji":"🎯","tokens_out":12671,"duration_ms":98277,"temperature":0.7,"pith_summary":"Quantum state discrimination asks how to tell apart nonorthogonal states, and any strategy must trade errors against inconclusive outcomes. This paper reports an optical experiment that realizes the optimal version of this trade-off — the minimal error rate for a fixed rate of inconclusive outcomes — for N=2, 3, 5, and 7 equally likely symmetric states of a qubit. The implementation uses a two-step recipe: a probabilistic state-separation map first increases the distinguishability of the inputs, and a minimum-error measurement is then applied to the successfully transformed outputs. The measured error rates follow the theoretical curve that interpolates between the minimum-error extreme (no inconclusive outcomes) and the maximum-confidence or unambiguous extreme (maximal inconclusive rate), demonstrating the two-step approach in a programmable optical setup that can be extended to higher dimensions.","feed_headline":"Experiment hits quantum-optimal error trade-off for up to 7 states","feed_subtitle":"Two-step optical method interpolates between minimum-error and unambiguous quantum-state discrimination.","key_machinery":"The central object is the two-step decomposition of the optimal FRIO measurement: a probabilistic quantum map called state separation, which with maximal success probability $p_s(\\theta')=(\\sin\\theta/\\sin\\theta')^2$ rotates the symmetric input states $|\\alpha_j(\\theta)\\rangle$ into the more distinguishable states $|\\beta_j(\\theta')\\rangle$, followed by the minimum-error POVM $\\hat\\Pi_k^{\\rm ME}=\\frac{2}{N}|u_k\\rangle\\langle u_k|$ on the successful outputs. The argument is carried by the identities $Q(\\theta')=p_f(\\theta')=1-p_s(\\theta')$ and the closed-form expression $P_{\\min}^e(Q)$ of Eq. (13), which together show that fixing the separation angle fixes the inconclusive rate and yields the optimal error rate. Experimentally, the machinery is a path-encoded qubit, a polarization ancilla, a programmable liquid-crystal spatial light modulator that implements the controlled unitary of Eq. (7), and a lens-focal-plane detector array that realizes the projective measurement of Eq. (15).","core_discovery":"The central claim is that optimal fixed-rate-of-inconclusive-outcomes (FRIO) discrimination — minimizing the error probability $P_e$ under a fixed rate $Q$ of inconclusive outcomes — can be implemented as a two-step process, and that an optical realization of that process matches the quantum-optimal curve for symmetric qubit states. For $N$ equally likely symmetric pure states, the paper derives and verifies the trade-off $$P_{\\min}^e(Q)=\\frac{1}{N}\\Big[(N-1)(1-Q)-\\sqrt{(1-Q)^2-(Q-Q_{\\rm MC})^2}\\Big],$$ which reduces to the minimum-error bound at $Q=0$ and to the optimal unambiguous ($N=2$) or maximum-confidence ($N>2$) rate at $Q=Q_{\\rm MC}=\\cos 2\\theta$. The experiment prepares path-encoded qubit states with a spatial light modulator, performs the separation by coupling the path modes to a polarization ancilla via a programmable liquid-crystal modulator, and completes the minimum-error step with a lens-focal-plane detector array that implements the quantum Fourier transform measurement. The reported data follow this curve closely for all four $N$, with agreement improving as $N$ grows.","pith_inferences":["If the depolarization of the transmissive liquid-crystal modulator at high gray levels is the dominant imperfection, then replacing that device with a phase-only modulator or adding active polarization correction should pull the high-$Q$ data points closer to the ideal curve; this is a testable upgrade of the present setup.","The two-step structure suggests the same decomposition could be used for optimal fixed-trade-off discrimination in non-symmetric state sets, provided a suitable probabilistic separation map is available, rather than only for the symmetric qubit case demonstrated here.","Because the paper compares its data against the ideal pure-state curve rather than the true optimum for the actual mixed states, the claim that the implementation closely reaches the minimum error rate is a conservative statement; recomputing the optimum for the measured visibilities $V_j$ and phases $\\phi_j$ would provide a stricter test."],"forward_implications":["The two-step recipe provides a constructive way to reach any point on the optimal error–inconclusive trade-off curve: set the separation angle $\\theta'$ to fix $Q$, then run the minimum-error measurement.","For quantum communication protocols that need to keep both the error rate and the inconclusive rate below the levels of the standard extreme strategies, the demonstrated scheme offers a tunable intermediate operating point controlled by a single programmable element.","Because the separation stage uses only a two-dimensional ancilla and the measurement stage works for any $N$, the same experimental platform can be extended to symmetric qudit states encoded in $d$ path modes.","The improvement of the agreement as $N$ increases supports the interpretation that experimental imperfections largely independent of $N$ are the main source of deviation."],"supporting_citations":[{"why":"Introduces the optimal FRIO measurement and supplies the theoretical trade-off curve that the experiment uses as its benchmark.","marker":"[20]"},{"why":"Provides the analytical solution for optimal discrimination with a fixed rate of inconclusive outcomes and the critical value of Q.","marker":"[21]"},{"why":"Introduces quantum state separation, the probabilistic map that forms the first step of the two-step implementation.","marker":"[34]"},{"why":"Gives the maximal success probability and measurement operators for parametric separation of symmetric pure states, used in the ancilla coupling.","marker":"[22]"},{"why":"Supplies the lens-camera method for high-dimensional minimum-error discrimination used as the second step.","marker":"[29]"},{"why":"Provides the minimum-error POVM and the maximum-confidence value for symmetric qudit states that underlie the theoretical analysis.","marker":"[36]"},{"why":"The prior two-state FRIO experiment that this work extends to N-state sets; it defines the baseline for the demonstration.","marker":"[33]"}],"fun_headline_variants":["Quantum-optimal error trade-off demonstrated for up to 7 qubit states","Two-step method achieves optimal state discrimination for qubits","Experiment verifies optimal error-inconclusive curve for N qubits","Up to 7 states: experiment reaches quantum-optimal discrimination"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The theoretical curve used as the benchmark assumes the separated states are exactly the ideal pure symmetric states, but the liquid-crystal modulator partially depolarizes the ancilla — an effect the paper's own footnote 5 acknowledges and leaves for future work — and introduces phase errors, especially at large separation angles, so the comparison against the ideal curve does not account for the true mixedness of the prepared states.","fun_headline_variants_meta":{"raw":{"variants":["Quantum-optimal error trade-off demonstrated for up to 7 qubit states","Two-step method achieves optimal state discrimination for qubits","Experiment verifies optimal error-inconclusive curve for N qubits","Up to 7 states: experiment reaches quantum-optimal discrimination"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00093,"raw_usage":{"total_tokens":3993,"prompt_tokens":969,"completion_tokens":3024,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":585,"completion_tokens_details":{"reasoning_tokens":2951}},"tokens_in":585,"tokens_out":3024,"duration_ms":23673,"temperature":1.0,"reasoning_tokens":2951,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T15:09:18.313194+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the true minimum error for the actual mixed states $\\hat\\rho_j(\\theta')$ of Eq. (18) using the measured visibilities $V_j$ and phases $\\phi_j$, and check the data at the largest separation angles ($\\theta'_6$ and $\\theta'_7$), where depolarization is strongest: if the experimental error rates lie above that recomputed optimum by more than the experimental uncertainty, the claim that the implementation closely reaches the minimum error rate for fixed $Q$ is falsified.","supporting_citations":[{"cited_title":"Mohseni, A","cited_arxiv_id":null,"evidence_quote":"Introduces quantum state separation, the probabilistic map that forms the first step of the two-step implementation."},{"cited_title":"Experimental optimal discrimination of $N$ states of a qubit with fixed rates of inconclusive outcomes","cited_arxiv_id":"2411.14537","evidence_quote":"Supplies the lens-camera method for high-dimensional minimum-error discrimination used as the second step."},{"cited_title":"G ´omez, E","cited_arxiv_id":null,"evidence_quote":"Provides the minimum-error POVM and the maximum-confidence value for symmetric qudit states that underlie the theoretical analysis."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"The prior two-state FRIO experiment that this work extends to N-state sets; it defines the baseline for the demonstration."}],"review_version":1}