{"id":"e2801bfc-19bc-483e-84a8-1c3693dc2b30","arxiv_id":"2608.09076","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"For qutrit GHZ and symmetric W states, the way entanglement splits after measuring one particle is determined by repeated-index versus all-different-index structure, not by the GHZ-versus-W label.","lead":"The paper computes exactly how qutrit versions of the GHZ and W states split their entanglement after one particle is measured, in both standard and mutually unbiased bases. It finds that for these states, the relevant dividing line is whether the state's index pattern repeats, not the usual GHZ-versus-W family label.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The hybrid/clean dichotomy is a computational-basis artifact: for a generic unadapted basis, |W_{p,p,q}> has outcome-independent Schmidt rank 2, so the abstract's 'repeated-index vs all-different' claim overreaches.","rationale":"The algebraic results in Appendices A-F appear internally consistent, and the paper is admirably explicit in Sec. 2.3.4 and Sec. 6 that the cabling dictionary is bookkeeping and that generic bases may behave differently. My concern is not with the calculations but with the abstract-level interpretation. Two points matter. First, the generic-basis calculation above shows the hybrid behavior is not intrinsic: a generic basis makes |W_{p,p,q}> outcome-independent with Schmidt rank 2 for every outcome, so the 'hybrid' classification is an artifact of the computational basis. Second, the stated 'repeated-index versus all-different-index bag structure' dichotomy cannot accommodate |GHZ3>, whose bag is repeated yet which is grouped with the clean states. These are scope and interpretation problems, not algebraic errors, so the correct response is to retain the conditional verdict and require the authors to restate the central claim in basis-relative terms. The reader's weakest_assumption identified the same core basis-dependence issue; my analysis sharpens it into an explicit counterexample and adds the GHZ classification problem, hence partial agreement. No change to the reader's verdict is needed.","tokens_in":28294,"tokens_out":18804,"duration_ms":358792,"concrete_test":"Use |W_{0,1,1}> and a generic, non-symmetry-adapted single-qutrit basis, e.g. a Haar-random SU(3) basis or any basis whose nine computational overlaps are all nonzero. For each basis vector |u_j>, project particle A onto |u_j>, normalize the residual state on B,C, and compute the Schmidt rank of the resulting 3x3 coefficient matrix. Repeat for |GHZ3> and |W_{0,1,2}>. Prediction from the algebra above: for every outcome j, rank(|W_{0,1,1}>)=2, rank(|GHZ3>)=3, and rank(|W_{0,1,2}>)=3, with no outcome-dependent switching. If confirmed, the hybrid/clean distinction is a computational-basis artifact and the abstract's structural claim must be revised.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The abstract's central claim that the relevant structural line is 'repeated-index versus all-different-index bag structure' is not supportable as stated. Table 3 itself places |GHZ3>, which has the repeated-index bag {a,a,a}, together with the all-different |W_{0,1,2}>, so the named dichotomy does not even classify GHZ. More decisively, the 'hybrid' outcome-dependence of |W_{p,p,q}> is an artifact of the computational basis. For a generic single-qutrit basis {|u_j>}, the branch after outcome j is proportional to u_j^{p*}(|pq>+|qp>) + u_j^{q*}|pp>. In the {|p>,|q>} basis its coefficient matrix is (1/N)[[u_q^*, u_p^*],[u_p^*, 0]], whose determinant is -(u_p^*)^2/N^2; whenever u_j has a nonzero p-component, which is true for every vector of a generic basis, the residual Schmidt rank is exactly 2 for every outcome. Thus the probability-weighted mixture of R=2 and R=1 in Table 2 is a property of measuring in a basis aligned with the bag labels, not a property of the state. Section 6 already concedes that all coefficient matrices are simple 'because of measuring in a basis naturally adapted to the permutation symmetry'; the generic-basis calculation shows the hybrid classification disappears entirely outside the computational basis. The abstract-level statement that the |W_{p,p,q}> states 'alone show probability-weighted, outcome-dependent behaviour' is therefore false as a statement about qutrit states; it holds only for computational-basis measurement. The conclusion should be restricted to the computational basis and similarly aligned bases, with the bag-structure vocabulary replaced by a basis-relative statement.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies measurement-induced entanglement splitting for tripartite qutrit states. For |GHZ_3>, for the six two-same-one-different symmetric W states |W^sym_{p,p,q}>, and for the all-different state |W_{0,1,2}>, it computes the residual bipartite Schmidt rank and the full Schmidt spectrum after a single-particle projective measurement, both in the computational basis and in the non-computational mutually unbiased bases. The algebraic results are collected in Tables 1, 2, 6, and 7 and derived in Appendices A--G. The paper then interprets the rank patterns with a two-strand cabling extension of Aravind's ring-and-link picture, in which the residual Schmidt rank R=1,2,3 is identified with 0,1,2 strands remaining linked. The authors explicitly state that this cabling dictionary is a labeling convention built to match the independently computed Schmidt rank, not a topological invariant. The paper's advertised structural conclusion is that the relevant divide is repeated-index versus all-different-index bag structure rather than GHZ-class versus W-class membership.","tokens_in":28622,"tokens_out":8367,"duration_ms":80290,"significance":"If the paper is read as a calculation for computational-basis and MUB measurements on permutation-symmetric qutrit GHZ and W states, the contribution is solid and useful. The appendices contain complete, checkable derivations; the MUB calculations correctly reduce to unit-modulus phases, and the reported eigenvalues, probabilities, and Schmidt ranks are internally consistent. The paper is also unusually candid about the status of its topological dictionary: Section 2.3.4 states plainly that no topological invariant is computed from the link diagrams, that the cabling choice n=d-1 is not singled out by any physical principle, and that the dictionary inherits basis dependence. This candor is a strength. The paper also identifies concrete experimental platforms where the Schmidt-rank predictions could be tested. The central interpretive claim, however, is broader than the evidence: the abstract and conclusion present the repeated-index/all-different-index distinction as the relevant structural line 'for qutrits', whereas the calculations only support that statement for two symmetry-adapted measurement bases.","major_comments":[{"comment":"The central claim that the |W^sym_{p,p,q}> states 'alone show probability-weighted, outcome-dependent behaviour' is not supported as a statement about qutrit states; it is a property of the computational-basis measurement. For a generic single-qutrit measurement basis {|u_j>}, the post-measurement branch for |W^sym_{p,p,q}> is proportional to u_j^{p*}(|pq>+|qp>) + u_j^{q*}|pp>. In the {|p>,|q>} basis its coefficient matrix is proportional to [[u_q^*, u_p^*],[u_p^*,0]], whose determinant is -(u_p^*)^2. Whenever u_j^p is nonzero, which is true for every outcome of a generic basis, the residual Schmidt rank is exactly 2 for every outcome. Thus the hybrid R=2/R=1 structure in Table 2 disappears outside the computational basis. Section 6 already concedes that every coefficient matrix is simple only 'because of measuring in a basis naturally adapted to the permutation symmetry', but the abstract and conclusion are not correspondingly scoped. The claims should be reworded to refer to computational-basis and MUB measurements on these symmetric states, not to qutrits in general.","section":"Abstract; Sec. 6 (Scope of the calculation)"},{"comment":"The paper's stated dichotomy, 'repeated-index versus all-different-index bag structure', does not classify the states in Table 3 as presented. The |GHZ_3> state has a repeated-index bag {a,a,a}, yet it is grouped with the all-different |W_{0,1,2}> as clean and outcome-independent, while the repeated-index |W^sym_{p,p,q}> states are classified as hybrid. Section 6 introduces an additional qualification ('The GHZ state is the special case, not the template'), but this qualification is absent from the abstract's slogan. Either the dichotomy must be refined to distinguish the triple-repeated GHZ bag from the two-same-one-different bags, or the claim must be explicitly restricted to the W-family. As written, the abstract's central structural line is not consistent with the paper's own table.","section":"Abstract; Table 3; Sec. 6"},{"comment":"The topological dictionary is explicitly only a bookkeeping device: the paper states that no linking number, braid-group representation, or polynomial invariant is computed from the diagrams and shown to reproduce R. This is acknowledged, and I do not treat it as a deceptive claim. However, the manuscript's framing as an 'extension' of Aravind's correspondence should not be allowed to obscure that the cabling dictionary has, by the authors' own account, no predictive or explanatory content beyond the Schmidt-rank calculation. The Conclusion already says this, and the Section 6 caveat that the dictionary 'should be read as a naming convention' is appropriate. I would ask the authors to carry this scoping into the abstract's first sentence, which currently says the paper 'shows' a division that matters 'for qutrits'.","section":"Sec. 2.3.4; Sec. 6"}],"minor_comments":[{"comment":"In Appendix E, the conjugate transpose M_j^dagger is written with the same entries as M_j; for complex r_j the (1,2) and (2,1) entries of M_j^dagger M_j should be r_j-bar/3 and r_j/3. The displayed determinant and eigenvalues are unaffected because the calculation uses |r_j|^2=1, but the notation should be corrected.","section":"Eqs. (125)--(128)"},{"comment":"The entries of the matrix N and of the rank-one matrix w_j w_j^dagger require complex conjugation in several off-diagonal positions; as printed, N appears not to equal I + w_j w_j^dagger for arbitrary unit-modulus s_j,t_j. The final eigenvalues {2/3,1/6,1/6} are nevertheless correct because the argument only needs the rank-one structure and the unit-modulus property.","section":"Appendix F, Eqs. (140)--(143)"},{"comment":"The text says 'we perform all projective measurements on qubit A' when the system under consideration is a qutrit. The word 'qubit' should be 'qutrit' here and in any analogous statements.","section":"Appendix B.1"},{"comment":"Minor language issues include 'Extending the Aravind's correspondence' and 'the Aravind's' should be 'Aravind's correspondence' or 'the Aravind correspondence'.","section":"Abstract and Sec. 1"},{"comment":"The author list in reference [20] is incorrectly punctuated ('B-G. Englert. I. Bengtsson and K. Zyczkowski'); it should be 'B.-G. Englert, I. Bengtsson, and K. Zyczkowski'.","section":"Reference [20]"}],"recommendation":"major_revision","confidential_remarks":"The algebraic content is sound and the appendices are a genuine resource, but the paper's advertised message requires re-scoping. The generic-basis calculation in the stress-test note is correct and directly undermines the abstract's 'for qutrits' claim; this is not merely a wording issue. If the authors restrict the structural claim to computational-basis and MUB measurements on the symmetric states treated here, and adjust the abstract and conclusion accordingly, I would view the paper as a solid contribution. If the broad claim is retained without a generic-basis analysis, the central assertion would not be supported."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Bottom line: the algebraic core is solid and the paper is honestly written, but the headline claim overreaches. The stress-test note is correct. For |W_{p,p,q}>, measure in a generic single-qutrit basis; the coefficient matrix has determinant -(u_p^*)^2/N^2, so every outcome has Schmidt rank 2. The probability-weighted mixture of R=2 and R=1 in Table 2 is a property of the computational basis, not of the state. The authors concede in Sec. 6 that the simple matrices arise because the bases are adapted to permutation symmetry, but the abstract still presents repeated-index versus all-different-index as 'the relevant structural line.' That line doesn't even classify GHZ, which has a repeated bag {a,a,a} yet behaves like the all-different state. So the conclusion needs to be restricted to CB and similarly adapted bases.\n\nWhat is genuinely new: exact residual Schmidt ranks and eigenvalues for all six two-same-one-different symmetric qutrit W states and the all-different W state under CB and MUB. I checked Appendices E and F; the p=1/3, eigenvalues (3±√5)/6 and {2/3,1/6,1/6}, and ranks R=2 and 3 follow from unit-modulus phases and are internally consistent. The GHZ CB/MUB results are standard. The paper is also admirably explicit that the two-strand cabling dictionary is bookkeeping, not a derived invariant; Sec. 2.3.4 says this plainly and lists what would be needed to close the gap. That kind of honesty should count.\n\nSoft spots in proportion: the abstract and Sec. 6 overstate the basis-relative pattern as a state property. The stress-test calculation is simple and should be added; it would force a more accurate scope statement. The all-different/repeated-index dichotomy also needs refinement to handle GHZ. The topological dictionary is not a flaw because it is openly labeled, but it is not evidence for anything beyond the Schmidt-rank tables.\n\nWho this is for: people working on qutrit entanglement measures or measurement-induced entanglement splitting will want these reference tables. The paper deserves a serious referee; the algebra is checkable and the overclaim is fixable with a revised abstract and a generic-basis appendix. I would not desk-reject it.","headline":"The Schmidt-rank tables are correct and worth having, but the abstract's repeated-index versus all-different-index claim is too strong: the hybrid behavior of the two-same-one-different W states is an artifact of measuring in a symmetry-adapted basis.","tokens_in":29193,"tokens_out":2275,"would_cite":true,"duration_ms":21973,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["03.67.-a","03.67.Mn"],"model":"deepseek-v4-flash","headline":"For qutrits, the way entanglement splits after a single-particle measurement is set by the bag structure of the labels—repeated-index versus all-different-index—not by whether the state belongs to the GHZ or W class.","keywords":["quantum entanglement","qutrits","GHZ states","W states","Schmidt rank","mutually unbiased bases","topological entanglement","knot theory"],"falsifier":"Measure one particle of |GHZ_3>, |W_{0,1,2}> and |W_{p,p,q}> in a basis that is neither the computational basis nor an MUB, compute the residual Schmidt ranks for every outcome, and check whether |W_{0,1,2}> still tracks |GHZ_3> while the |W_{p,p,q}> family shows mixed ranks; a generic basis that breaks the repeated-index/all-different-index distinction would falsify the paper's central claim. A purely mathematical falsifier would be computing a genuine link invariant (e.g., a Jones polynomial or braid trace) of the proposed cabled diagrams and checking whether it changes value in step with R, which the authors state has not been established.","tokens_in":28027,"feed_emoji":"🔗","tokens_out":12101,"duration_ms":86447,"temperature":0.7,"pith_summary":"Multipartite entanglement is usually discussed in two families, GHZ and W, with different behaviour expected under measurement. The paper asks whether that division is the one that matters for tripartite qutrits when a single particle is measured, and answers no. By direct calculation in the computational basis and in every mutually unbiased basis, the authors show that the states split into two classes by the bag structure of their labels: the all-different state |W_{0,1,2}> behaves exactly like |GHZ_3>, with uniform outcome probabilities and an outcome-independent residual Schmidt rank in each basis, while the six repeated-index states |W_{p,p,q}> are the only ones showing probability-weighted, outcome-dependent ranks. The paper packages this result in a two-strand cabling extension of the ring-and-link picture, and states plainly that the cabling dictionary is a naming convention built on independently computed Schmidt ranks, not a topological invariant.","feed_headline":"All-different qutrit W state splits like GHZ under measurement","feed_subtitle":"For tripartite qutrits, repeated-index vs all-different bags, not GHZ/W class, set the residual Schmidt rank.","key_machinery":"The argument runs on the coefficient matrix M of the post-measurement two-qutrit state: the residual Schmidt rank R is the number of non-zero eigenvalues of M†M, so every conclusion reduces to a 3×3 or 2×2 matrix analysis. To express the three possible values of R, the paper replaces each topological ring with a cable of n=d−1=2 parallel strands, mapping R=1,2,3 to 0, 1, or 2 linked strands after the measurement cut. The second measurement context is supplied by the complete set of four mutually unbiased bases for a qutrit, built from powers of the cube root of unity, whose defining uniform overlap forces the outcome-uniform probabilities. The classification itself is organized by permutation bags of labels, repeated-index versus all-different-index, where a bag is the multiset of three indices whose distinct permutations are superposed to form the symmetric state.","core_discovery":"The central discovery is that, for permutation-symmetric tripartite qutrit states, the way entanglement splits after a single-particle measurement is fixed by whether the computational labels form a repeated-index bag {p,p,q} or an all-different bag {0,1,2}, not by GHZ-class versus W-class membership. |GHZ_3> under the computational basis leaves a separable pair (Schmidt rank R=1, spectrum {1,0,0}) for every outcome, and under any mutually unbiased basis leaves a maximally entangled pair (R=3, spectrum {1/3,1/3,1/3}). The all-different W state |W_{0,1,2}> reproduces this clean pattern: R=2 with spectrum {1/2,1/2,0} for every computational outcome, and R=3 with spectrum {2/3,1/6,1/6} in any MUB. The six |W_{p,p,q}> states behave differently only in the computational basis: outcome p (probability 2/3) yields a maximally entangled qubit-like pair with R=2, outcome q (probability 1/3) yields a product state with R=1, and the third outcome never occurs; any MUB gives uniform outcomes with R=2 and unequal spectrum {(3+√5)/6, (3−√5)/6, 0}.","pith_inferences":["A direct experimental check is available: prepare |GHZ_3>, |W_{0,1,2}> and one |W_{p,p,q}> in photonic or trapped-ion qutrit platforms, measure in the computational basis and in an MUB, and verify the predicted probabilities and residual Schmidt ranks.","If a generic-basis measurement preserves the repeated-index/all-different-index distinction, the natural generalization for qudits of dimension d is that partition structure of the label multiset plays the organizing role, with (d−1)-strand cables replacing two-strand ones.","The hybrid 'predominant link plus contextual exception' language for |W_{p,p,q}> under computational-basis measurement is an operational description of a probability-weighted mixture of ranks, and could be reinterpreted as a statement about the entanglement distillable from the residual pair without any topological reading."],"forward_implications":["In the computational basis, the six |W_{p,p,q}> states are the only states studied whose residual rank is outcome-dependent: outcome p (probability 2/3) leaves R=2, outcome q (probability 1/3) leaves R=1.","|W_{0,1,2}> matches |GHZ_3> in both measurement contexts—uniform probabilities and outcome-independent ranks—so the GHZ/W family line does not organize measurement-induced splitting for qutrits.","In any of the three mutually unbiased bases, every state studied gives uniform outcome probabilities; the |W_{p,p,q}> family stays at R=2 with unequal spectrum, while |W_{0,1,2}> reaches R=3.","The two-strand cabling dictionary maps R=1,2,3 to 0, 1, or 2 linked strands, which gives a consistent vocabulary for all seven states across two bases, but it remains a labeling convention until a link invariant is shown to track R."],"supporting_citations":[{"why":"Supplies the ring-and-link correspondence, rings for particles and cutting for measurement, that the paper extends from qubits to qutrits.","marker":"[1]"},{"why":"Develops the topological-entanglement framework and the cutting operation that the qutrit extension builds on.","marker":"[10]"},{"why":"Establishes the three-qubit GHZ-versus-W inequivalence that the paper argues is not the relevant divide for qutrit splitting.","marker":"[14]"},{"why":"Gives the authors' earlier entanglement-splitting classification for tripartite states that this work generalizes to qutrits.","marker":"[16]"},{"why":"Introduces the link-fluidity vocabulary for symmetric Dicke states that informs the hybrid predominant/contextual classification.","marker":"[17]"},{"why":"Provides the mutually-unbiased-bases construction, d+1 bases for prime dimension d, used for all non-computational measurements.","marker":"[19]"},{"why":"Demonstrates the basis-dependence of the qubit ring-and-link correspondence, the caveat the qutrit dictionary inherits.","marker":"[28]"},{"why":"Shows photonic GHZ entanglement beyond qubits, the platform the paper names for testing its Schmidt-rank predictions.","marker":"[29]"},{"why":"Maps the larger structure of three-qutrit entanglement classes that bounds the scope of the bag-structure claim.","marker":"[30]"}],"fun_headline_variants":["Index bag, not GHZ or W, decides qutrit split","Qutrit split rule: index pattern, not state family","Qutrit entanglement: repeated vs distinct indices split","Measurement splits qutrits by index bag, not class","Qutrit entanglement split: index bag trumps class"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The claim rests on the assumption that the clean-versus-hybrid pattern seen in the computational basis and in the mutually unbiased bases reflects an intrinsic property of the states; the paper itself notes that every coefficient matrix it obtains is diagonal or rank-one-plus-identity precisely because those bases are adapted to the permutation symmetry, and that a generic unadapted basis should not be expected to preserve the pattern.","fun_headline_variants_meta":{"raw":{"variants":["Index bag, not GHZ or W, decides qutrit split","Qutrit split rule: index pattern, not state family","Qutrit entanglement: repeated vs distinct indices split","Measurement splits qutrits by index bag, not class","Qutrit entanglement split: index bag trumps class"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001447,"raw_usage":{"total_tokens":5981,"prompt_tokens":1248,"completion_tokens":4733,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":864,"completion_tokens_details":{"reasoning_tokens":4649}},"tokens_in":864,"tokens_out":4733,"duration_ms":28653,"temperature":1.0,"reasoning_tokens":4649,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T00:00:52.745014+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure one particle of |GHZ_3>, |W_{0,1,2}> and |W_{p,p,q}> in a basis that is neither the computational basis nor an MUB, compute the residual Schmidt ranks for every outcome, and check whether |W_{0,1,2}> still tracks |GHZ_3> while the |W_{p,p,q}> family shows mixed ranks; a generic basis that breaks the repeated-index/all-different-index distinction would falsify the paper's central claim. A purely mathematical falsifier would be computing a genuine link invariant (e.g., a Jones polynomial or braid trace) of the proposed cabled diagrams and checking whether it changes value in step with R, which the authors state has not been established.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the ring-and-link correspondence, rings for particles and cutting for measurement, that the paper extends from qubits to qutrits."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Develops the topological-entanglement framework and the cutting operation that the qutrit extension builds on."},{"cited_title":"Bhattacharyya and S","cited_arxiv_id":null,"evidence_quote":"Gives the authors' earlier entanglement-splitting classification for tripartite states that this work generalizes to qutrits."},{"cited_title":"Bhattacharyya and S","cited_arxiv_id":null,"evidence_quote":"Introduces the link-fluidity vocabulary for symmetric Dicke states that informs the hybrid predominant/contextual classification."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the mutually-unbiased-bases construction, d+1 bases for prime dimension d, used for all non-computational measurements."},{"cited_title":"Erhard, M","cited_arxiv_id":null,"evidence_quote":"Shows photonic GHZ entanglement beyond qubits, the platform the paper names for testing its Schmidt-rank predictions."},{"cited_title":"Chitambar, R","cited_arxiv_id":null,"evidence_quote":"Maps the larger structure of three-qutrit entanglement classes that bounds the scope of the bag-structure claim."}],"review_version":1}