{"id":"25677b96-4b9e-4705-8d40-1cb7fc32bd80","arxiv_id":"2608.13376","paper_version":1,"verdict":"ACCEPT","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"One fixed stabilizer protocol distills an exact CCZ state from six copies of any unknown pure non-stabilizer qubit state, with optimal success probability M3^lin(psi)/3.","lead":"Six copies of any unknown pure non-stabilizer qubit state can be converted by one fixed stabilizer circuit into an exact CCZ magic state, a basic resource for universal quantum computation. The optimal success probability equals one third of the linearized order-three stabilizer Rényi entropy of the input, giving this entropy a direct operational meaning.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"No significant objection identified; the six-copy optimality proof survives scrutiny, and the only non-elementary step (Clifford 3-design in Prop. 1) is secondary and standard.","rationale":"The reader's weakest assumption correctly identifies the Clifford 3-design use in Proposition 1 as the least elementary point in the paper, and I agree that the k≤9 upper bound would fail if the commutant contained extra operators. However, that is not load-bearing for the central claim, which is Theorem 1 (six-copy threshold and optimal probability). That theorem is proven by a separate, target-specific argument using polynomial ideals, stabilizer nullity, and stabilizer purity P2, none of which require the 3-design. I verified the key steps: the dimension of M_k for k<6 is zero, the six-copy protocol's success probability is exactly M3^lin/3, the upper bound via reduction to |A6⟩ and Lemma 11 is internally consistent, and the ancilla reduction in Lemma 11 is justified because the measured Pauli acts trivially on the stabilizer ancilla. The eight-copy protocol is more intricate but is a strengthening rather than the core existence threshold; its proof is sufficiently explicit (Krawtchouk coefficients and Clifford symmetries) and, in any case, the six-copy protocol already implies Corollary 3. I therefore find no load-bearing defect and recommend the verdict remain unchanged, while noting that an independent check of Lemma 11 would further harden the central optimality proof.","tokens_in":27859,"tokens_out":33319,"duration_ms":349917,"concrete_test":"Independently re-derive Lemma 11 by exhaustive computation on the 4-qubit state |η⟩: enumerate all 4-qubit Pauli operators P, compute the post-measurement states for both outcomes, and verify (i) no no-type-(iii) branch can be Clifford-equivalent to |χ⟩⊗|ζ⟩ by comparing Pauli spectra, and (ii) the only surviving type-(iii) branch has total success probability 2/3; this would close the final gap in the six-copy optimality proof.","verdict_should_be":"UNCHANGED","load_bearing_attack":"I checked the central claim (six copies necessary and sufficient, with optimal success probability M3^lin/3) and found no load-bearing defect. The polynomial reduction in Theorem 12 is sound: any successful branch has c_r(a,b) vanishing on the six stabilizer rays, so c_r = γ_r f_6; the step to K_r restricted to the symmetric subspace being proportional to |χ⟩⟨A_6| follows because ψ^⊗6 spans Sym_6. Lemma 11 then correctly forces exactly one type-(iii) Pauli measurement from the nullity budget (4 → 3+ν(ζ)) and rules out the other outcome by showing all post-measurement Pauli expectations are 0 or ±1, making the output a stabilizer state. The ancilla reduction to |η⟩ is justified because any Pauli commuting with the stabilizers of a stabilizer state acts as ±1 on that state. The only non-elementary step I could identify is in the secondary upper bound of Proposition 1, where the single-qubit Clifford group being a unitary 3-design is used to conclude A_k = λ_k Π_M_k for 6 ≤ k ≤ 9. This is a standard theorem (Zhu, PRA 96, 062336), and even if this step failed, the six-copy threshold, the optimal six-copy success probability, and the universality corollary would remain intact. I therefore do not regard the 3-design step as a load-bearing concern for the paper's central claim.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies exact magic state concentration: a fixed stabilizer protocol that, given k copies of an unknown pure non-stabilizer qubit state, outputs an exact target magic state with state-dependent success probability. The main result is that for the CCZ target, six copies are necessary and sufficient. The authors give an explicit six-copy protocol with success probability M3^lin(psi)/3, prove that this is optimal among all universal six-copy protocols, and show that fewer copies give zero success probability. They also construct an eight-copy protocol with success probability 2M3^lin(psi)/3 and prove a general upper bound for 6 <= k <= 9 showing that the success probability for any fixed non-stabilizer target is at most a constant times M3^lin(psi). Block repetition gives asymptotic rates R(psi -> CCZ) >= M3^lin/12, and an upper bound R <= C M3^lin log(1/M3^lin) with C < 1.81, so the scaling is optimal up to logarithmic factors. A corollary is that any unknown pure non-stabilizer qubit state enables universal quantum computation by exact CCZ injection with a fixed protocol.","tokens_in":28162,"tokens_out":10720,"duration_ms":105465,"significance":"The result is significant because it gives an exact operational interpretation to the linearized order-three stabilizer Renyi entropy as the optimal success probability at the minimal copy number, and it establishes a qualitative difference between CCZ and T targets: exact T-state concentration is known to be obstructed, whereas exact CCZ concentration is possible from any non-stabilizer pure state. The proofs are explicit and checkable, with parameter-free derivations: the overlap identity (B9), the reduction of the eight-copy filter to a Krawtchouk polynomial calculation (B38)-(B46), and the reduction of the optimality theorem to a detailed analysis of the state |A6> in Theorem 12. The central formulas express the success probability through the previously defined monotone M3^lin with no fitting parameters. The paper also strengthens Reichardt's universality result by removing the state-dependent adaptation of the protocol, replacing it with a fixed stabilizer procedure whose only state dependence is the success probability.","major_comments":[],"minor_comments":[{"comment":"The operator Omega-tilde_6 is called a projector, but on the full six-qubit Hilbert space it is not idempotent; it becomes a projector only when restricted to the symmetric subspace, as is clear from the appendix. This should be stated explicitly to avoid confusion in the proof sketch.","section":"Main text, Eq. (7)"},{"comment":"The inequality P2(zeta) >= 2/3 for any single-qubit pure state is used without proof or citation; it follows directly from the Bloch-sphere constraint x^2+y^2+z^2=1, but a brief justification would make the appendix more self-contained.","section":"Appendix C.1, Lemma 11"},{"comment":"The step invoking the Clifford 3-design is terse: the intertwining relation (C36) is what transfers the commutant statement from Sym^{k-6}(C^2) to M_k, and making this explicit would help the reader verify the bound for k=9.","section":"Appendix C.2, Prop. 3"},{"comment":"The statement that 0 <= M3^lin(psi) <= 4/9 for a single qubit, with equality at the Bravyi-Kitaev T-type states, is stated without derivation or reference; a short argument or citation would be helpful.","section":"Main text, after Eq. (2)"}],"recommendation":"accept","confidential_remarks":"My own reading aligns with the reader's report. The central identities are explicit and checkable, the optimality proof for k=6 is thorough, and the Clifford 3-design step used for the k<=9 upper bounds is standard and does not affect the six-copy threshold. The minor points listed are presentation issues only; I see no need for further technical revision before publication."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear colleague,\n\nThis is a solid, well-proved result. The authors show that six copies of any unknown pure non-stabilizer qubit state can be converted, by a fixed stabilizer protocol, into one exact CCZ state with success probability M3^lin(psi)/3, and that this is optimal among all universal six-copy stabilizer protocols. They also give an eight-copy protocol with success probability 2/3 M3^lin(psi), an upper bound for up to nine copies showing that the scaling is optimal, and an asymptotic rate analysis. The six-copy threshold is new and resolves a clean open question.\n\nWhat I like: the proofs are explicitly checkable. The key polynomial step—that any successful branch must have coefficient proportional to f6—is exactly right, and the reduction from A6 to chi with the nullity argument is careful. I checked the main steps of Theorem 12 and found no gap. The paper also does a useful job of placing the result inside the Clifford commutant framework, which makes the connection to stabilizer testing transparent. The eight-copy protocol is a real construction with a Krawtchouk reduction that is clean to follow.\n\nThe soft spots are real but not damaging. The setting is pure, exact, and noiseless; this is acknowledged, and the authors do not oversell fault-tolerance implications. The k<=9 upper bound in Proposition 1 relies on the Clifford group being a unitary 3-design. That is a standard theorem, but it is a different kind of tool from the elementary Pauli algebra used elsewhere. If that step failed, the main six-copy threshold would remain intact—the upper bound there is proved by a separate target-specific argument that the stress test confirmed. So I do not treat the 3-design step as load-bearing for the paper's central claim. The eight-copy protocol is not known optimal, and the authors say so. The asymptotic upper bound is loose by a log factor, but the point is scaling.\n\nOverall: this paper should get a serious referee. The referee should focus on the appendices, especially the Clifford 3-design step and the proof of Lemma 11, because those are the most compressed. I would take it to reading group and would cite it for the M3^lin operational interpretation.\n\nRecommendation: accept with attention to the 3-design step; it is a real paper.","headline":"A clean, well-proved six-copy threshold for exact CCZ concentration from unknown pure magic states, with M3^lin as the operational success probability; worth serious refereeing.","tokens_in":28711,"tokens_out":2004,"would_cite":true,"duration_ms":19397,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["81P68","81P45"],"pacs":["03.67.-a","03.67.Pp"],"model":"deepseek-v4-flash","headline":"Six copies of any unknown pure non-stabilizer qubit state are necessary and sufficient to distill one exact CCZ state, with optimal success probability set by the linearized order-three stabilizer Rényi entropy.","keywords":["magic state concentration","CCZ state","stabilizer Rényi entropy","exact distillation","Clifford group","symmetric subspace","universal quantum computation"],"falsifier":"To settle the claim, one can compute the full Clifford commutant on the symmetric subspace $M_k$ for $6\\le k\\le 9$ and check whether any operator besides the projector $\\Pi_{M_k}$ commutes with all Clifford unitaries; finding one would break the claimed bound. Alternatively, run the six-copy protocol on a known non-stabilizer state and test whether the acceptance frequency equals $\\tfrac{1}{3}M_3^{\\mathrm{lin}}(\\psi)$ as the input is varied.","tokens_in":27650,"feed_emoji":"⚛️","tokens_out":13274,"duration_ms":120833,"temperature":0.7,"pith_summary":"Magic states are the non-Clifford resources that lift stabilizer circuits to universality, but standard distillation assumes prior knowledge of the input, such as a noise model or proximity to a target. This paper shows that for pure qubit states this assumption can be dropped: a fixed protocol built from Clifford gates, Pauli measurements, and classical feedforward converts six copies of any unknown non-stabilizer state $|\\psi\\rangle$ into one exact three-qubit controlled-controlled-Z (CCZ) magic state, with success probability $\\tfrac{1}{3} M_3^{\\mathrm{lin}}(\\psi)$. Six copies are also necessary: no universal stabilizer protocol can produce any fixed non-stabilizer output from five or fewer copies. The linearized order-three stabilizer Rényi entropy $M_3^{\\mathrm{lin}}$ thus acquires a concrete operational meaning: it sets the optimal probability of concentration at the minimal block size, and it controls the state dependence of any protocol up to nine input copies. If the claim is right, every pure non-stabilizer qubit state enables universal quantum computation by exact CCZ injection with one fixed procedure, the only unknown being the overhead.","feed_headline":"Six copies of unknown magic states make one exact CCZ","feed_subtitle":"With six copies of any unknown magic qubit, one fixed protocol yields a universal gate—no input alignment or tomography.","key_machinery":"The load-bearing object is the stabilizer-orthogonal symmetric subspace $M_k$, defined as the subspace of $k$-qubit symmetric states orthogonal to $|s\\rangle^{\\otimes k}$ for every single-qubit stabilizer state $|s\\rangle$. In the polynomial representation of the symmetric subspace, a state lies in $M_k$ exactly when its associated degree-$k$ homogeneous polynomial vanishes on the six vertices of the single-qubit stabilizer octahedron; the minimal obstruction is $f_6(a,b)=ab(a^4-b^4)$, and the identity $M_3^{\\mathrm{lin}}(\\psi)=6|f_6(a,b)|^2$ ties the entropy to this geometry. The protocols measure $X^{\\otimes k}$ and $Z^{\\otimes k}$ and postselect on syndromes that project onto $M_k$. For $k=6$, $M_6$ is one-dimensional, so the postselected state is independent of the input; a short chain of Pauli measurements and Clifford corrections converts that unique state into $|\\mathrm{CCZ}\\rangle$ with probability $2/3$, giving the overall factor $1/3$. Clifford invariance of $M_k$ is what makes the protocol universal and state-independent.","core_discovery":"The paper's central discovery is a sharp six-copy threshold for exact universal magic-state concentration. For every pure non-stabilizer qubit state $|\\psi\\rangle$, there is a single stabilizer protocol $\\Lambda_6$ that consumes $|\\psi\\rangle^{\\otimes 6}$ and, on success, outputs an exact $|\\mathrm{CCZ}\\rangle$ state; the success probability is $\\Pr_{\\Lambda_6}(\\psi^{\\otimes 6}\\to\\mathrm{CCZ}) = \\tfrac{1}{3}M_3^{\\mathrm{lin}}(\\psi)$. Among all universal six-copy stabilizer protocols that output CCZ, this probability is optimal, and for $k<6$ no protocol succeeds. The same entropy governs the broader landscape: for $6\\le k\\le 9$, the success probability of exact conversion into any fixed non-stabilizer state is at most $\\tfrac{7(k-5)}{2(k+1)}M_3^{\\mathrm{lin}}(\\psi)$, so the linear scaling in $M_3^{\\mathrm{lin}}$ is best possible in that range. An eight-copy protocol doubles the six-copy probability to $\\tfrac{2}{3}M_3^{\\mathrm{lin}}(\\psi)$. As a corollary, a fixed stabilizer procedure turns any unknown pure non-stabilizer qubit state into universal quantum computation through exact CCZ injection, with only the overhead depending on the state.","pith_inferences":["Beyond the paper's own claims, the subspace geometry suggests that no seven-copy block can improve on the six-copy success probability, while at ten copies new Clifford-invariant operators should enter and may unlock success probabilities not tied solely to $M_3^{\\mathrm{lin}}$.","A natural testable extension is to qudits: replacing $f_6$ by the lowest-degree polynomial vanishing on the local stabilizer octahedron would predict the minimal copy number in each dimension, and the protocol's success probability should be governed by the corresponding stabilizer entropy.","On near-term hardware, the protocol's accepted branch is essentially a stabilizer test, so the predicted acceptance probability $\\tfrac{1}{3}M_3^{\\mathrm{lin}}(\\psi)$ can be checked by Pauli expectation measurements without tomography of the input state."],"forward_implications":["A single fixed stabilizer protocol makes every pure non-stabilizer qubit state sufficient for universal quantum computation by exact CCZ injection; repetition supplies CCZ states at finite expected cost.","With access to one T-state catalyst, the six- and eight-copy protocols yield exact catalytic conversions to two T states, with success probabilities $\\tfrac{1}{3}M_3^{\\mathrm{lin}}(\\psi)$ and $\\tfrac{2}{3}M_3^{\\mathrm{lin}}(\\psi)$.","Block repetition of the eight-copy protocol achieves asymptotic concentration rates at least $\\tfrac{1}{12}M_3^{\\mathrm{lin}}(\\psi)$ for CCZ and $\\tfrac{1}{6}M_3^{\\mathrm{lin}}(\\psi)$ for catalytic T; the matching upper bound shows this scaling is optimal up to logarithmic factors.","The linearized order-three stabilizer Rényi entropy becomes an operationally defined quantity: the optimal probability of extracting one exact CCZ state from six unknown copies.","The same measurement structure applies to any mixed state supported on the symmetric subspace for $k=6,8$, producing an exact CCZ state on success with probability proportional to the state's weight on $M_k$."],"supporting_citations":[{"why":"Foundational magic-state distillation template; defines the T-type target and the syndrome-based conversion approach that this paper makes universal.","marker":"[3]"},{"why":"Shows every pure non-stabilizer state enables universal computation with stabilizer operations, with the protocol possibly depending on the input state.","marker":"[8]"},{"why":"Extends state-dependent distillation universality and sets the baseline this paper improves to a fixed protocol.","marker":"[9]"},{"why":"Provides the deterministic catalyzed conversion of one CCZ state into two T states used to derive catalytic T-state concentration.","marker":"[23]"},{"why":"Defines stabilizer Rényi entropies and their linearized forms, giving $M_3^{\\mathrm{lin}}$ its basic properties.","marker":"[31]"},{"why":"Supplies the six-copy Clifford-invariant operator and stabilizer-testing threshold that the six-copy protocol mirrors.","marker":"[33]"},{"why":"Proves exact CCZ-to-T conversion under stabilizer operations is impossible, explaining why CCZ rather than T is the natural exact target.","marker":"[41]"},{"why":"Gives the relative-entropy-of-magic bounds and additivity used for the asymptotic rate upper bound.","marker":"[43]"},{"why":"Establishes that the Clifford group is a unitary 3-design, the fact behind the nine-copy commutant upper bound.","marker":"[47]"}],"fun_headline_variants":["Six copies of any magic qubit make one exact CCZ","Optimal six-copy magic concentration yields exact CCZ","Universal gate from six copies of unknown magic states","Six-copy threshold: exact CCZ from any pure magic state","No tomography needed: six copies to a universal gate"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the Clifford group's action on the relevant symmetric subspaces has no hidden extra symmetries for six to nine copies; if such hidden structure existed, the claimed $M_3^{\\mathrm{lin}}$ upper bounds could be circumvented, though the six-copy impossibility below six is proved separately and would survive.","fun_headline_variants_meta":{"raw":{"variants":["Six copies of any magic qubit make one exact CCZ","Optimal six-copy magic concentration yields exact CCZ","Universal gate from six copies of unknown magic states","Six-copy threshold: exact CCZ from any pure magic state","No tomography needed: six copies to a universal gate"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000329,"raw_usage":{"total_tokens":1912,"prompt_tokens":1101,"completion_tokens":811,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":717,"completion_tokens_details":{"reasoning_tokens":730}},"tokens_in":717,"tokens_out":811,"duration_ms":8292,"temperature":1.0,"reasoning_tokens":730,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T12:31:20.918924+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"To settle the claim, one can compute the full Clifford commutant on the symmetric subspace $M_k$ for $6\\le k\\le 9$ and check whether any operator besides the projector $\\Pi_{M_k}$ commutes with all Clifford unitaries; finding one would break the claimed bound. Alternatively, run the six-copy protocol on a known non-stabilizer state and test whether the acceptance frequency equals $\\tfrac{1}{3}M_3^{\\mathrm{lin}}(\\psi)$ as the input is varied.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Shows every pure non-stabilizer state enables universal computation with stabilizer operations, with the protocol possibly depending on the input state."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the deterministic catalyzed conversion of one CCZ state into two T states used to derive catalytic T-state concentration."},{"cited_title":"Gross, S","cited_arxiv_id":null,"evidence_quote":"Gives the relative-entropy-of-magic bounds and additivity used for the asymptotic rate upper bound."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Establishes that the Clifford group is a unitary 3-design, the fact behind the nine-copy commutant upper bound."}],"review_version":1}