{"id":"2bbd3927-407f-49ee-a2f6-9e39be120afe","arxiv_id":"2412.10297","paper_version":3,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"For any even d, an exchange-symmetrized qudit Bell basis exists, and linear-evolution local-measurement devices can distinguish 2d-1 of its Bell states, the maximum allowed.","lead":"The authors construct a complete set of entangled two-qudit states, for any even dimension d, in which each state has a definite symmetry when the particles' internal states are swapped. They show a linear-optics measurement can reliably tell apart 2d-1 of these states, which is the largest number allowed, and outline a qudit dense coding protocol.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Basis construction and the 2d−1 distinguishability protocol are sound, but the claimed saturation of a 2d−1 upper bound for arbitrary even d is inherited from Ref [32] and supported only by a heuristic counting argument; if that bound does not extend, the abstract's optimality claim loses support.","rationale":"The reader's conditional verdict identifies exactly the soft spot: the upper bound is inherited rather than proved for arbitrary even d. I agree with that assessment. The constructive parts of the paper are verifiable and appear correct: the round-robin pairing gives disjoint correlation classes, the DFT phases give orthogonality within each class, and the HOM bunching/antibunching argument plausibly distinguishes the chosen 2d−1 codewords. The paper contains no fitted parameters and no machine-checked proofs, but the explicit basis is a real mathematical construction. The main risk is not in the construction but in the optimality statement, which is central to the abstract's claim that the result 'achieves the upper bound in general.' A concrete counterexample search for d=6 would settle whether the inherited bound genuinely extends. In the absence of such a test, a conditional verdict is the appropriate level of confidence. I see no reason to move the verdict to reject or unverified, because the new basis and the 2d−1 distinguishability result stand even if the upper-bound generalization were to fail. No ad hominem is intended; the issue is an unproven generalization, not an internal inconsistency or a flaw in the basis construction itself.","tokens_in":11980,"tokens_out":17790,"duration_ms":777398,"concrete_test":"For d=6, numerically search over passive linear-optical unitaries U in U(12) and over subsets of the paper's 36 Bell states for a set of 12 states whose two-photon outcome supports under number-resolving detection are pairwise disjoint, a necessary condition for unambiguous LELM discrimination. If such a set is found, the claimed 2d−1 upper bound is false; if an extensive random search finds none, the inherited bound is corroborated for d=6. An independent analytical check would be to re-derive the maximum number of disjoint outcome supports for exchange-symmetrized inputs without invoking Ref [32] and verify that it equals 2d−1 for all even d.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim has two parts: the explicit exchange-symmetrized qudit Bell basis, and the statement that LELM can distinguish 2d−1 Bell states and that this is the upper bound. The basis construction in Eqs. (7)–(9) is internally sound: orthogonality within a correlation class follows from the DFT phase choice, opposite-parity states are orthogonal from the (−1)^p factor, and different correlation classes have disjoint pair supports. The odd-d no-go argument is also correct. The load-bearing concern is the saturation clause. In the section 'Maximal distinguishability of bell states for even d', the upper bound k = 2d−1 is taken from Ref [32], which treated d=2^n hyperentangled qubit Bell states, and is extended to arbitrary even d by the statement that the argument 'applies equally for any d provided the Bell basis is symmetrized under particle exchange.' The supporting reasoning is heuristic: it counts 'first' and 'second' particle detections across 2d channels and subtracts one for bosonic exchange symmetry. This counting does not account for the fact that a Bell state generally has a support consisting of many possible two-photon outcome patterns, not a single detection pair; the total number of two-photon outcome patterns in 2d modes is d(2d+1), so a simple 2d-channel count cannot by itself establish a 2d−1 upper bound without additional structure. Therefore the optimality clause of the abstract is not independently proven in this paper, even though the positive result that 2d−1 states can be distinguished appears correct.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper constructs a complete exchange-symmetrized generalized Bell basis for pairs of qudits of even dimension d, using a round-robin pairing of single-particle basis states into correlation classes and d/2-point DFT phase factors to generate phase classes. It proves that no such complete basis can exist for odd d via a unitary coefficient-matrix argument. The authors further claim that a linear-optical LELM device (a 50/50 beam splitter followed by generalized polarizing beam splitters) can unambiguously distinguish 2d−1 states from this basis for any even d, and that this saturates the upper bound for LELM-based Bell-state discrimination. A dense-coding protocol with 2d−1 codewords is sketched.","tokens_in":12269,"tokens_out":21708,"duration_ms":193075,"significance":"The exchange-symmetrized basis construction is a clean and correct generalization of the hyperentangled qubit bases to arbitrary even dimensions, and the odd-d impossibility proof is elegant and sound. The proposed distinguishability scheme, if fully substantiated, would be practically relevant for high-dimensional quantum communication with linear optics, and the dense-coding application is a natural use case. However, the optimality claim is currently supported only by a heuristic extension of a result proved for d=2^n, which is a load-bearing gap given the paper's title and abstract emphasis on Bell-state distinguishability.","major_comments":[{"comment":"The paper asserts in the section 'Maximal distinguishability of bell states for even d' that the upper bound k = 2d−1, previously established for d=2^n in Ref. [32], applies 'equally for any d provided the Bell basis is symmetrized under particle exchange.' This extension is not proved. The accompanying counting argument—that a detection signature consists of a pair of detections across 2d channels and that exchange symmetry reduces the count by one—does not suffice: the total number of two-photon detection patterns in 2d modes is d(2d+1), which is much larger than 2d−1, so the bound cannot follow from this counting alone. Since the abstract's claim 'achieves the upper bound in general' depends on this gap, the authors should either supply a rigorous derivation for arbitrary even d or explicitly present the upper bound as a conjecture or as an inherited result rather than as a theorem proven here.","section":"Maximal distinguishability of bell states for even d"},{"comment":"The positive statement that the device in Figure 1 can unambiguously distinguish 2d−1 Bell states is not demonstrated in the text. The paper does not define the set of states being distinguished (the dense-coding section later suggests the set {c=0,p=0} ∪ {c≠0,p=0,1}), nor does it show that the post-beam-splitter detection supports of these states are pairwise disjoint. The generalized HOM effect gives only the bunching/antibunching bit; one still needs to show that the round-robin pairing prevents any overlap of detection patterns between different correlation classes and that the chosen phase classes are resolved. An explicit mapping from states to detection signatures (e.g., a table for a small example) would make the claim verifiable.","section":"Maximal distinguishability of bell states for even d and Figure 1"}],"minor_comments":[{"comment":"Equations (8) and (11) are garbled in the typeset text, with the fragment 'j p 2 k' appearing where a mathematical expression is intended. The intended phase appears to be φ_{j,p} = (4πj/d)⌊p/2⌋, using the d/2-point DFT; please correct the typesetting and define the floor notation.","section":"Symmetric Basis for Even Dimension, Eqs. (8) and (11)"},{"comment":"The phrase 'This achieves the upper bound in general' is ambiguous because the paper treats only even d, and the d=3 case has a different upper limit; please specify 'for even d' and clarify the relationship to the d=3 result.","section":"Abstract"},{"comment":"In the discussion of the Walsh-matrix construction, the statement that phase classes can be built from [H(2^{n−1})]_{j,p} should specify that the column index is ⌊p/2⌋, since H(2^{n−1}) has only d/2 columns while p ranges over d values.","section":"Connection to hyperentangled basis"},{"comment":"The c=0 correlation class contains only symmetric states, despite the (−1)^p factor in Eq. (9), because each term |2j⟩|2j⟩ and |2j+1⟩|2j+1⟩ is individually symmetric under the exchange of internal states; a clarifying sentence would prevent readers from misinterpreting the odd-p states as antisymmetric.","section":"Symmetric Basis for Even Dimension, Eq. (9)"},{"comment":"The conclusion states that 'we have shown that an LELM device can unambiguously distinguish 2d−1 of these exchange-symmetrized qudit Bell states', but the body does not contain an explicit proof of this statement; either add the proof (as suggested in the major comments) or soften the conclusion to match the present level of detail.","section":"Conclusion"}],"recommendation":"major_revision","confidential_remarks":"The paper's contribution is split between a solid basis construction and an insufficiently supported distinguishability optimality claim. Since one of the authors is a co-author of Ref. [32], the manuscript should explicitly separate what is proven here from what is inherited from the d=2^n case, to avoid any appearance of overclaiming. The missing proof of the upper bound for arbitrary even d is the main obstacle; if it cannot be supplied, the abstract and title should be adjusted to claim only the construction and the 2d−1 codeword scheme, with optimality presented as a conjecture."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The thing to know: the core construction is real and the odd-d proof is clean. The paper builds a complete exchange-symmetrized Bell basis for any even d using round-robin pairings and DFT phases, and proves no such basis exists for odd d. The orthogonality and completeness arguments are simple and I did not find a gap. The paper is clearly written, includes a worked d=6 example, and draws a nice connection to Walsh matrices for d=2^n. That part is worth taking seriously.\n\nThe headline LELM claim splits in two. The positive result—that a scheme can unambiguously distinguish 2d−1 states for any even d—looks right: the HOM bunching/antibunching argument maps parity into spatial information, and the distinguishability counting matches the construction. The optimality part is not established. The 2d−1 upper bound is taken from Ref. [32], which treated d=2^n hyperentangled qubit Bell states, and is extended here to arbitrary even d by a heuristic argument: 2d detection channels, first and second detections, subtract one for bosonic exchange. That counting ignores the fact that a Bell state has support on many two-photon outcome patterns, not a single detection pair; the total number of two-photon patterns in 2d modes is d(2d+1), so the channel count does not by itself prove a 2d−1 bound. The authors assert the Ref. [32] argument 'applies equally for any d' but do not re-derive it. Since one author is shared, this is not circular, but it is an unproven extension of a published result. The abstract's 'achieves the upper bound in general' overstates what is shown.\n\nThe dense coding application is only a sketch; it does not verify the full detection signature analysis for the chosen codewords, but it is explicitly presented as an outline.\n\nNet: the basis construction alone justifies the paper, and the LELM distinguishability scheme is a useful addition. The optimality clause should either be proven for all even d or softened to a conjecture. I would send this to a serious referee, with the explicit request to focus on whether the 2d−1 upper bound extends beyond d=2^n. I would cite the basis construction in my own work and bring the paper to reading group.","headline":"A genuinely new exchange-symmetrized qudit Bell basis for even d, with a correct odd-d no-go proof; the LELM optimality claim, however, rests on an unproven extension of a prior bound.","tokens_in":12869,"tokens_out":2464,"would_cite":true,"duration_ms":21984,"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":"This paper constructs a complete exchange-symmetrized qudit Bell basis for every even dimension d and proves an LELM device can distinguish 2d−1 of its states, matching the upper bound.","keywords":["qudit Bell basis","exchange symmetry","Bell-state measurement","LELM distinguishability","Hong-Ou-Mandel effect","dense coding","round-robin pairing","generalized Bell states"],"falsifier":"A concrete disproof would be an explicit LELM circuit for some even d that unambiguously distinguishes 2d Bell states, which would violate the claimed ceiling; alternatively, a direct calculation for d = 4 showing that the generalized Hong-Ou-Mandel bunching/antibunching signatures of two distinct symmetric states overlap, so that one of the purported 2d−1 distinguishable classes is not actually resolved, would collapse the saturation claim.","tokens_in":11745,"feed_emoji":"⚛️","tokens_out":2784,"duration_ms":29678,"temperature":0.7,"pith_summary":"The paper introduces a generalized Bell basis for pairs of qudits whose shared dimension d is any even number, with every basis state an eigenstate of exchanging the two particles' internal states. It proves that no such complete symmetrized basis exists for odd d, and that for even d an LELM device can unambiguously distinguish 2d−1 states of this basis. That number matches the known upper bound for Bell-state distinguishability under linear evolution and local measurement, previously established only for d equal to a power of two or three. The result matters because LELM devices are the experimentally practical route to Bell measurements, and reaching the bound for all even d widens the range of qudit communication protocols that can run without conditional logic.","feed_headline":"Qudit Bell bases reach the 2d−1 LELM limit for every even d","feed_subtitle":"Exchange-symmetrized Bell states push linear-optics Bell measurement to its ceiling for all even qudit dimensions.","key_machinery":"The central object is an exchange-symmetrized qudit Bell basis built from two ingredients: correlation classes supplied by a round-robin tournament pairing of the d single-particle basis states, and phase classes supplied by a discrete Fourier transform of size d/2 that makes states within a class mutually orthogonal while preserving definite swap symmetry. The generalized Hong-Ou-Mandel effect at a 50/50 beam splitter is what carries the distinguishability argument, separating symmetric (bunching) from antisymmetric (antibunching) states and reducing the LELM discrimination task to counting distinct two-particle detection signatures.","core_discovery":"For any even single-particle dimension d, the authors build d² fully entangled, mutually orthogonal qudit Bell states that are simultaneous eigenstates of the swap operator, with (d+1)d/2 symmetric and (d−1)d/2 antisymmetric states. The construction separates each state into a correlation class, given by a pairing of the d basis states according to a round-robin tournament schedule, and a phase class, given by a discrete Fourier transform over the d/2 pairs. No complete exchange-symmetrized Bell basis exists for odd d, because antisymmetric states require skew-symmetric coefficient matrices, and every odd-dimensional skew-symmetric matrix has zero determinant. Using the symmetrized basis, the paper shows that a device built from a 50/50 beam splitter and generalized polarizing beam splitters distinguishes 2d−1 Bell states under LELM, saturating the general upper bound. The key mechanism is the generalized Hong-Ou-Mandel effect: symmetric input states bunch on one side of the beam splitter while antisymmetric input states antibunch, converting the phase-class parity into a spatial signature.","pith_inferences":["Editorial inference: the same round-robin construction may extend to other exchange-invariant tasks, such as symmetric or antisymmetric subspace projections, where the tournament pairing gives a natural orthogonal decomposition of the two-qudit Hilbert space.","Editorial inference: the paper's distinguishability result relies on the upper bound of Ref. [32] transferring unchanged to these new bases; a direct check of whether the 2d−1 bound can be beaten by an unsymmetrized basis for some even d would delimit how much the symmetry assumption actually costs.","Editorial inference: a concrete experimental prediction is that, in a photonic realization with path-encoded qudits, the 2d−1 distinguishable states should appear as exactly 2d−1 distinct two-photon detection signatures, with the antisymmetric states always antibunching and the symmetric states always bunching.","Editorial inference: the dense-coding sketch uses only 2d−1 of the d² basis states; whether adaptively choosing different subsets of the basis can boost the rate beyond one bit per transmitted qudit in a noisy setting is a natural next question."],"forward_implications":["An LELM device can unambiguously distinguish 2d−1 Bell states for every even d, extending the known limits beyond d = 2, 3, and powers of two.","This gives a qudit dense-coding protocol with 2d−1 codewords: Alice encodes her message by locally transforming the shared state into one of the distinguishable Bell states, and Bob decodes with the passive LELM apparatus.","For odd d, no complete exchange-symmetrized Bell basis exists, so reaching the same distinguishability ceiling for odd d would require a different, unsymmetrized construction.","The basis provides a family of fully entangled states with definite swap symmetry, which can serve as resources for quantum communication and algorithmic protocols whose qubit versions rely on exchange symmetry of the Bell states.","Because the phase factors can be chosen from Walsh matrices instead of a DFT when d is a power of two, the construction reproduces the earlier hyperentangled symmetrized bases as a special case."],"supporting_citations":[{"why":"Supplies the 2d−1 upper bound on LELM-distinguishable Bell states and the generalized Hong-Ou-Mandel bunching/antibunching argument, which the paper adapts to arbitrary even d.","marker":"[32]"},{"why":"Provides the prior exchange-symmetrized hyperentangled basis for d = 2^n that this paper generalizes, and the definition of fully entangled bipartite states used throughout.","marker":"[33]"},{"why":"Establishes the d = 2 benchmark that only three of four Bell states can be distinguished by LELM, the base case for the general bound.","marker":"[24, 25]"},{"why":"Provides the previously known d = 3 maximal distinguishability result that the present even-d result complements.","marker":"[34]"},{"why":"Supplies the round-robin tournament construction used to generate the d−1 non-trivial correlation classes of the new basis.","marker":"[59]"},{"why":"Justifies the restriction of the optimal device to no auxiliary modes by showing auxiliary photons cannot increase the number of distinguishable Bell states in a projective LELM measurement.","marker":"[71, 72]"}],"fun_headline_variants":["Exchange-symmetrized Bell basis reaches the 2d−1 LELM ceiling","Even qudit dimensions hit the 2d−1 Bell-state limit","No complete exchange-symmetrized Bell basis for odd d","Qudit Bell states saturate the LELM bound for even d","Exchange-symmetrized Bell basis distinguishes 2d−1 states for any even d"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the 2d−1 upper bound on LELM-distinguishable Bell states, argued in Ref. [32] for hyperentangled qubit states with d = 2^n, holds unchanged for arbitrary even d and for the new exchange-symmetrized basis.","fun_headline_variants_meta":{"raw":{"variants":["Exchange-symmetrized Bell basis reaches the 2d−1 LELM ceiling","Even qudit dimensions hit the 2d−1 Bell-state limit","No complete exchange-symmetrized Bell basis for odd d","Qudit Bell states saturate the LELM bound for even d","Exchange-symmetrized Bell basis distinguishes 2d−1 states for any even d"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001131,"raw_usage":{"total_tokens":4711,"prompt_tokens":970,"completion_tokens":3741,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":586,"completion_tokens_details":{"reasoning_tokens":3635}},"tokens_in":586,"tokens_out":3741,"duration_ms":26467,"temperature":1.0,"reasoning_tokens":3635,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T16:00:04.964022+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A concrete disproof would be an explicit LELM circuit for some even d that unambiguously distinguishes 2d Bell states, which would violate the claimed ceiling; alternatively, a direct calculation for d = 4 showing that the generalized Hong-Ou-Mandel bunching/antibunching signatures of two distinct symmetric states overlap, so that one of the purported 2d−1 distinguishable classes is not actually resolved, would collapse the saturation claim.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the 2d−1 upper bound on LELM-distinguishable Bell states and the generalized Hong-Ou-Mandel bunching/antibunching argument, which the paper adapts to arbitrary even d."},{"cited_title":"Bouwmeester, J.-W","cited_arxiv_id":null,"evidence_quote":"Provides the prior exchange-symmetrized hyperentangled basis for d = 2^n that this paper generalizes, and the definition of fully entangled bipartite states used throughout."},{"cited_title":"Karimipour, A","cited_arxiv_id":null,"evidence_quote":"Supplies the round-robin tournament construction used to generate the d−1 non-trivial correlation classes of the new basis."}],"review_version":1}