{"id":"a34af9e6-c535-4854-9850-0c22fee97431","arxiv_id":"2608.01011","paper_version":1,"verdict":"ACCEPT","confidence":"HIGH","novelty_score":8.0,"correctness_risk":"low","formal_verification":"none","parameter_count":1,"one_line_summary":"AME(7,d) states exist iff d is at least 3; explicit constructions cover all odd d and all d congruent to 2 modulo 4, including the previously open dimensions 6 and 10.","lead":"The paper proves that a seven-party quantum state that is maximally entangled across every split exists exactly when each party has dimension at least 3, and gives explicit constructions for every allowed dimension. It settles the last open cases in a well-known classification problem for absolutely maximally entangled states, which are used in quantum error correction, secret sharing, and holographic models.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"No significant objection identified","rationale":"The reader accepted with high confidence, and my pass agrees. I checked the determinant certificates in Eq. (5), the Fano/coset support analysis leading to Eq. (13), the null-vector image-exclusion certificates in Table II, and the case split in Table III. The only point not fully proved in the text is the odd-dimension AME theorem from Ref. [8], but the precise condition used (full row rank of Gamma_{S,B}, equivalently an invertible 3x3 minor) follows from a two-line character-orthogonality computation valid for every q, with the determinants being units for every odd q. I found no hidden assumption in the coupled construction: the binary and odd sectors are correctly combined, the cross term is cancelled on all cuts (non-Fano via R=0, Fano via T=0), and normalization is correct. Hence no load-bearing concern lands; the reader's verdict stands.","tokens_in":7486,"tokens_out":43197,"duration_ms":442106,"concrete_test":"Independently verify the odd-dimension branch for non-prime-power q (e.g., q=15 and q=21) by directly computing the 35 three-party reduced density matrices of |G_q> from Eq. (4) and checking each equals I_{q^3}/q^3; this settles the only external dependency (Theorem 3 of Ref. [8]) without relying on the citation.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's central claim is supported by explicit, checkable constructions. The odd-dimension branch uses the full-row-rank condition from Theorem 3 of Ref. [8]; for the five orbit representatives the relevant 3x3 minors have determinants -1, 1, 2, -1, 1, all units for odd q, and the rank condition is a direct character-orthogonality argument that does not require q to be prime. The twice-odd construction is self-contained: off-diagonal binary blocks vanish except on the seven Fano triples (Eq. (13) and Table I), and the odd-sector off-diagonal blocks are killed on those seven cuts by the null-vector certificates in Table II (Eqs. (25)-(29)). I do not see an internally inconsistent step or a missing case in the split d = 2^alpha m. The only external dependency is the cited theorem, and the specific use made of it is easily verified; I do not regard this as a load-bearing weakness.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proves the complete existence classification of seven-partite absolutely maximally entangled states: AME(7,d) exists if and only if d ≥ 3. The proof is constructive. For every odd q ≥ 3, the cyclic quadratic-phase state |G_q> in Eq. (4) is shown to be AME by verifying, through the five translation representatives S = 012, 013, 014, 015, 024, that suitable 3×3 submatrices Γ_{S,R} have determinants −1, 1, 2, −1, 1, all units over Z_q for odd q. For d = 2q with q odd, the coupled state |Ψ_{2q}> in Eq. (31) combines the two logical states of the Steane code with the odd-dimensional states |G_0^{(q)}> and |G_1^{(q)}>. The binary cross-block R^S_{01} is shown to vanish on all 28 non-Fano three-subsets by a support argument using the simplex/Hamming code structure, and the odd-sector cross-block T^S_{01} is shown to vanish on the seven Fano cuts via the null-vector certificates in Table II. Together with the known nonexistence at d = 2, the known AME(7,4) state, the Reed–Solomon construction for powers of two, and the product property for AME states, the claimed classification follows.","tokens_in":7718,"tokens_out":18360,"duration_ms":200310,"significance":"If correct, this resolves previously open cases including AME(7,6) and AME(7,10), and completes the existence classification of AME states for all n ≤ 7. The constructions are explicit and checkable: the five submatrix determinants are computed, the Fano-cut cancellation is certified by the seven explicit null vectors in Table II, and Eq. (40) gives an explicit wavefunction for the twice-odd dimensions. The proof is largely self-contained, with only standard code-theoretic facts and one cited rank criterion used as inputs. The completeness of the case split in terms of d = 2^α m with m odd is clear. This is a clean and significant resolution of a well-known open problem.","major_comments":[],"minor_comments":[{"comment":"The proof of Eq. (6) relies entirely on Theorem 3 of Ref. [8], whose exact hypotheses are not restated. Since the classification requires composite odd dimensions (q = 9, 15, ...), please either quote the theorem's hypotheses explicitly or include the short character-orthogonality argument showing that the stated rank condition over Z_q is sufficient for every odd q. The argument is elementary, but as written the reader cannot verify the scope of the cited theorem.","section":"Odd local dimensions, Eq. (6)"},{"comment":"The notation 'I23 23' in Eq. (10) is ambiguous; it should be written as I_{2^3}/2^3 (equivalently I_8/8).","section":"Twice-odd local dimensions, Eq. (10)"},{"comment":"Typo: 'Accorrding' should be 'According'.","section":"Twice-odd local dimensions, before Eq. (34)"},{"comment":"The phrase 'the distance of C and (C+1) are both 4' could be misread, since C+1 contains words of weight 3. Please clarify that the minimum distance between distinct elements of the coset C+1 is 4 (as a nonlinear code), not the minimum Hamming weight of its elements.","section":"Twice-odd local dimensions, Eqs. (10) and (13)"}],"recommendation":"minor_revision","confidential_remarks":"The result appears correct and the manuscript is well within the standards of the journal. The only substantive request is to make the hypotheses of the cited rank criterion explicit, since the entire odd-dimension branch depends on it. The acknowledgment of GPT-5.6 assistance is unusual, but the authors state that they independently verified the argument; I do not see this as a correctness concern."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The headline: the paper proves AME(7,d) exists iff d≥3, and it does. The genuinely new piece is the coupled binary-odd construction for d=2q (odd q≥3), which covers all previously open dimensions 6, 10, 14, and so on. The cyclic quadratic-phase construction for every odd d is a welcome uniform family, though it leans on a cited theorem rather than being fully self-contained.\n\nWhat the paper does well is make the proof concrete. The odd branch reduces to five 3x3 determinants (-1, 1, 2, -1, 1), all units for odd q, so the rank condition is transparent. The twice-odd branch is worked out in real detail: binary cross blocks vanish on 28 non-Fano cuts via a clean support argument, and the seven Fano cuts are killed by explicit null-vector certificates in Table II. The case split d = 2^α m with α = 0, 1, 2, ≥3 covers every dimension, and the product property is standard. I also checked for circularity: nothing in the proof assumes the classification it is proving.\n\nSoft spots are minor. The main external dependency is Theorem 3 of Ref. [8], which gives the sufficient condition for quadratic-phase states to be AME; the paper does not restate its full hypotheses. A referee will need to confirm the theorem applies as stated, but the specific use is only rank checks on 3x3 matrices, so verification is cheap. The d=2 nonexistence and the product property are cited standard results, which is fine. The disclosure that GPT-5.6 assisted in developing the twice-odd construction is handled honestly: the authors state they independently verified everything, and the paper supplies enough detail for an independent check. I do not see that as a red flag.\n\nThis paper is for people working on AME states, quantum MDS codes, and multipartite entanglement. It closes a known open problem with explicit constructions, and the completeness of the classification is genuinely useful for anyone querying which AME(7,d) exist. It is not a field-reshaping breakthrough, but it is a solid, complete result. The construction techniques, especially the binary-odd coupling, may carry over to other small n.\n\nRecommendation: yes, send it to peer review. The referee should verify the hypotheses of Theorem 3 and spot-check the determinant and Table II computations, but the core argument is sound and the contribution is real.","headline":"Settles the last open existence question for AME(7,d) with a genuinely new construction family; the proof is explicit and checkable, and the central argument holds up.","tokens_in":693,"tokens_out":1602,"would_cite":true,"duration_ms":38209,"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":"Seven-qudit AME states exist exactly in local dimension $d\\ge 3$.","keywords":["absolutely maximally entangled states","AME(7,d)","seven qudits","quantum MDS codes","quadratic-phase states","Steane code","Fano plane","Hamming code"],"falsifier":"Build $|G_q\\rangle$ from Eq. (4) for an odd composite, say $q=9$, using exact arithmetic, and compute each of the 35 three-party reduced density matrices $\\rho_S=\\operatorname{Tr}_{S^c}|G_q\\rangle\\langle G_q|$; the odd-dimensional branch is false if any $\\rho_S\\neq I/9^3$. A cheaper check is to compute the five representative determinants in Eq. (5) modulo $q$ and confirm, via translation invariance, that every one of the 35 regenerated $\\Gamma_{S,R}$ has nonzero determinant modulo $q$.","tokens_in":7397,"feed_emoji":"⚛️","tokens_out":9803,"duration_ms":105047,"temperature":0.7,"pith_summary":"This paper settles the existence problem for seven-partite absolutely maximally entangled states: an AME(7,d) state exists if and only if $d\\ge 3$. The only negative case is $d=2$, already known, and the genuinely new work supplies explicit states in every odd dimension and every dimension congruent to 2 modulo 4. The odd-dimensional family is a cyclic quadratic-phase state built from the complement of the seven-cycle, and the twice-odd family couples the Steane-code logical basis to such phase states so that every three-party reduced density is maximally mixed. A reader should care because AME states are pure quantum MDS codes and are used in quantum secret sharing, parallel teleportation, and holographic error correction; closing the seven-party case removes an outstanding entry from the standard AME table.","feed_headline":"Seven-qudit entanglement exists for every d≥3","feed_subtitle":"Explicit states cover odd and twice-odd dimensions, leaving d=2 as the only impossible case.","key_machinery":"Two explicit state families carry the proof. The odd family is the cyclic quadratic-phase state $|G_q\\rangle$ with phase matrix the upper-triangular part of the complement of the seven-cycle; the AME check becomes invertibility of $3\\times3$ submatrices over $\\mathbb{Z}_q$. The twice-odd family is the coupled state $|\\Psi_{2q}\\rangle=(|0_L\\rangle|G_0^{(q)}\\rangle+|1_L\\rangle|G_1^{(q)}\\rangle)/\\sqrt{2}$, where $|0_L\\rangle,|1_L\\rangle$ are Steane-code logical states and $|G_a^{(q)}\\rangle$ are one-qudit phase-shifted copies. The central mechanism is the complementary vanishing of the binary cross block (nonzero only on the seven Fano triples $L_t=\\{t,t+1,t+3\\}$) and the odd cross block (vanis","core_discovery":"The paper claims the complete classification AME(7,d) exists iff $d\\ge 3$. For odd $q$, it constructs $|G_q\\rangle=q^{-7/2}\\sum_{x\\in\\mathbb{Z}_q^7}\\omega_q^{Q(x)}|x\\rangle$ with $Q(x)=x\\tilde\\Gamma x^T$ and $\\Gamma$ the complement of the seven-cycle; by a cited criterion, checking the state is AME reduces to showing each of 35 three-by-three submatrices $\\Gamma_{S,R}$ is invertible over $\\mathbb{Z}_q$, which translation symmetry reduces to five determinant computations. For $d=2q$, it forms $|\\Psi_{2q}\\rangle=(|0_L\\rangle|G_0^{(q)}\\rangle+|1_L\\rangle|G_1^{(q)}\\rangle)/\\sqrt{2}$, identifying the qubit-qudit basis with a qudit of dimension $2q$. The binary cross term $R_{01}^S$ vanishes on al","pith_inferences":["One can view the construction as transferring the obstruction from the binary sector to a small exceptional set of cuts; this suggests a general recipe for AME$(n,2q)$ whenever the binary code has dual distance $\\lfloor n/2\\rfloor+1$ and the partner phase state kills the exceptional cuts.","The complete seven-party classification makes the qubit case $d=2$ look like the sole number-theoretic exception; it is natural to test the same binary-odd coupling on the next open case, AME(8,4).","Replacing $\\mathbb{Z}_q$ in the phase state with other finite abelian groups, if the invertibility criterion carries over, would give a direct construction for even dimensions without invoking the product property, possibly simplifying the $\\alpha\\ge2$ branches.","Because AME(7,d) is a pure $[[7,0,4]]_d$ MDS code, the classification also closes the existence question for seven-qudit QMDS codes; the remaining hard cases shift to lengths $n\\ge8$."],"forward_implications":["For every $d\\ge 3$, the paper gives explicit normalized pure states: Eq. (4) for odd $d$, Eq. (31) for $d\\equiv2\\pmod4$, and product or Reed-Solomon states for the remaining powers of two.","Every seven-partite system with local dimension $d\\ge3$ admits a pure state maximally entangled across every bipartition, so the seven-party AME table is complete.","The previously open cases $\\mathrm{AME}(7,6)$ and $\\mathrm{AME}(7,10)$ are resolved, with explicit states written out in Eq. (40).","Since AME(7,d) is equivalent to a pure quantum MDS code $[[7,0,4]]_d$, the classification proves such codes exist exactly for $d\\ge3$.","The failure at $d=2$ is isolated: no other local dimension obstructs seven-party absolutely maximally entangled states."],"supporting_citations":[{"why":"Provides Theorem 3, the quadratic-phase-state AME criterion used to verify every odd dimension.","marker":"[8]"},{"why":"Proves nonexistence of AME(7,2), the only negative case in the classification.","marker":"[9]"},{"why":"Defines the Steane-code logical states whose support structure is used in the binary sector of the coupled state.","marker":"[15]"},{"why":"Supplies the uniform projection property of the Hamming code used to show the diagonal three-party marginal is maximally mixed.","marker":"[16]"},{"why":"Gives the AME product property used to combine the odd and power-of-two factors.","marker":"[17]"},{"why":"States the product property for AME states, cited with [17] when extending to products of dimensions.","marker":"[18]"},{"why":"Supplies the known AME(7,4) state and the generalized Reed-Solomon construction covering powers of two.","marker":"[19]"},{"why":"Provides the known AME(7,4) state used for the $\\alpha=2$ case in the decomposition $d=2^\\alpha m$.","marker":"[20]"},{"why":"Gives the generalized Reed-Solomon code background for the minimal-support AME(7,$2^\\alpha$) states.","marker":"[21]"}],"fun_headline_variants":["Seven-qudit AME states exist iff d ≥ 3","Complete classification: AME(7,d) exists for all d≥3","Maximal entanglement in seven qudits: only d=2 fails","All dimensions ≥3 host seven-party maximally entangled states","Existence proof: AME(7,d) iff d≥3, d=2 impossible"],"cache_read_input_tokens":2816,"weakest_assumption_plain":"The load-bearing premise is the cited Theorem 3 of Ref. [8] that a quadratic-phase state over $\\mathbb{Z}_q^7$ is AME whenever, for every three-subset $S$, some $3\\times3$ submatrix $\\Gamma_{S,R}$ is invertible over $\\mathbb{Z}_q$; the paper does not prove this theorem or restate its full hypotheses, so every odd-dimension state (and every composite dimension built from it) depends on it.","fun_headline_variants_meta":{"raw":{"variants":["Seven-qudit AME states exist iff d ≥ 3","Complete classification: AME(7,d) exists for all d≥3","Maximal entanglement in seven qudits: only d=2 fails","All dimensions ≥3 host seven-party maximally entangled states","Existence proof: AME(7,d) iff d≥3, d=2 impossible"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000885,"raw_usage":{"total_tokens":3663,"prompt_tokens":751,"completion_tokens":2912,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":495,"completion_tokens_details":{"reasoning_tokens":2815}},"tokens_in":495,"tokens_out":2912,"duration_ms":26183,"temperature":1.0,"reasoning_tokens":2815,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T00:37:12.187277+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Build $|G_q\\rangle$ from Eq. (4) for an odd composite, say $q=9$, using exact arithmetic, and compute each of the 35 three-party reduced density matrices $\\rho_S=\\operatorname{Tr}_{S^c}|G_q\\rangle\\langle G_q|$; the odd-dimensional branch is false if any $\\rho_S\\neq I/9^3$. A cheaper check is to compute the five representative determinants in Eq. (5) modulo $q$ and confirm, via translation invariance, that every one of the 35 regenerated $\\Gamma_{S,R}$ has nonzero determinant modulo $q$.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides Theorem 3, the quadratic-phase-state AME criterion used to verify every odd dimension."},{"cited_title":"Higuchi and A","cited_arxiv_id":null,"evidence_quote":"Proves nonexistence of AME(7,2), the only negative case in the classification."},{"cited_title":"Rajchel-Mieldzio´ c, R","cited_arxiv_id":null,"evidence_quote":"Defines the Steane-code logical states whose support structure is used in the binary sector of the coupled state."},{"cited_title":"Ball and R","cited_arxiv_id":null,"evidence_quote":"Supplies the uniform projection property of the Hamming code used to show the diagonal three-party marginal is maximally mixed."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"States the product property for AME states, cited with [17] when extending to products of dimensions."},{"cited_title":"Rains, IEEE Transactions on Information Theory45, 1827 (1999)","cited_arxiv_id":null,"evidence_quote":"Supplies the known AME(7,4) state and the generalized Reed-Solomon construction covering powers of two."},{"cited_title":"Shi, M.-S","cited_arxiv_id":null,"evidence_quote":"Provides the known AME(7,4) state used for the $\\alpha=2$ case in the decomposition $d=2^\\alpha m$."},{"cited_title":"Raissi, A","cited_arxiv_id":null,"evidence_quote":"Gives the generalized Reed-Solomon code background for the minimal-support AME(7,$2^\\alpha$) states."}],"review_version":1}