{"id":"849cf245-a573-4e25-b95e-cc9368652a01","arxiv_id":"2610.00464","paper_version":1,"verdict":"ACCEPT","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"XC^(k)_d stabilizer states are exactly uniform superpositions over affine subspaces with diagonal phases one Clifford-hierarchy level above the stabilizers, giving Clifford-level realizations of type-III twisted quantum doubles and deterministic logical magic state preparation.","lead":"This paper builds a new class of quantum stabilizer codes from Pauli-X operators combined with diagonal gates drawn from a fixed level of the Clifford hierarchy, and proves that their unique code states are exactly uniform superpositions over affine subspaces with phases one level higher than the stabilizers.","discovery_kind":"first_principles","skeptic_critique":{"model":"deepseek/deepseek-v4-flash","headline":"Conjecture V.4's level-p lower bound for type-I/II cocycles rests on an explicitly assumed generalization of Haah's Wilson-line argument to twisted quantum doubles; the proven classification (Thm. IV.8) and type-III realization are unaffected, so the ACCEPT verdict stands.","rationale":"The reader's weakest_assumption precisely identifies the Wilson-line generalization step in Conjecture V.4, and my independent reading finds no additional load-bearing concern in the proven parts. I checked the main classification theorem IV.8: the affine-subspace support, the coset construction, and the level-shift argument via Cor. A.9 all follow without hidden assumptions, and the type-III realization in Lemma V.3 is derived explicitly with the correct degree-2 phase function (m=1, deg 2 → level 2). The magic-state protocol uses a valid counting argument (support sizes N_m not powers of p). The only true vulnerability is the unproven lower bound for type-I/II cocycles, which the authors themselves label a conjecture and openly state is unsupported. Since the paper's accepted claims are independent of that conjecture, and the conjecture is honestly confined, the ACCEPT verdict remains appropriate. I agree with the reader that no conditional or reject adjustment is warranted.","tokens_in":46478,"tokens_out":18510,"duration_ms":116393,"concrete_test":"Take the twisted quantum double D^ω(Z_3^2) with a single type-I cocycle (ω_I(a,b,c)=exp(2πi a_1 C(b_1,c_1)/3)) on a small triangulated torus, and construct the anyon whose topological spin is a primitive 9th root (existence guaranteed by Lem. D.16). Form the Wilson-line operator as a product of on-site A^ω_g(v) stabilizers along a noncontractible loop, following the method of Ref. [26] unmodified. Compute the phase acquired by a ground-state representative and compare it to the anyon's topological spin. If the phase differs from the known topological spin, Haah's procedure does not generalize to TQDs, invalidating the only support for Conjecture V.4's second case. If it matches, the conjecture gains support; either way the proven results stand.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's headline 'surprising consequence' — that non-abelian type-III phases are realizable at lower stabilizer level (k=2) than abelian type-I/II phases (k=p) — is entirely contingent on Conjectures I.2 and V.4. The second case (type-I/II require level p) has no lower-bound proof; the authors state this honestly in Sec. V around [b370]–[b371] and the conjectural step is explicitly flagged as 'Now assume that the same procedure generalizes to TQDs' ([b386]). That procedure is Haah's Wilson-line extraction of topological spin from products of on-site stabilizers, proven only for translation-invariant Pauli stabilizer codes (Ref. [26]). If this generalization fails, the p^2-root argument for type-I/II collapses, and the surprising consequence evaporates. However, this conjecture does not bear on the proven structural classification of XC(k) stabilizer states (Theorem IV.8) nor on the proved type-III level-2 realization (Lemma V.3). The paper's central proven content is internally coherent and consistent with cited anchors (k=1 and p=2 special cases, the carry-degree computation, the deterministic magic-state counting). The weakness is real but confined to a self-labeled conjecture.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek/deepseek-v4-flash","summary":"The paper introduces the XC^(k)_d stabilizer formalism for prime-dimensional qudits, in which stabilizers are products of Pauli-X strings with diagonal operators drawn from level k of the Clifford hierarchy. Section IV proves the central classification (Theorem IV.8): an n-qudit state is the unique codeword of an XC^(k)_d code iff, up to a permutation of qudits, it is a uniform superposition over an affine-linear subspace |x, Mx+c> carrying a diagonal phase D[φ] at level k+1 of the Clifford hierarchy; Algorithm 1 converts any such stabilizer subgroup into that canonical form. Section V shows that every twisted quantum double D^ω(Z_p^n) with ω a product of type-III cocycles and p an odd prime admits an XC^(2) codestate realization (Lemma V.3), and states Conjectures I.2/V.4 that cocycles with type-I or type-II factors require level-p stabilizers. Section VI gives a deterministic protocol that prepares a logical magic state in a pair of qudit toric codes by measuring XC^(k)_d stabilizers, improving on the probabilistic qubit protocol of Ref. [34]. Appendix D characterizes when an untwisted quantum double D(G) equals a twisted abelian quantum double D^ω(A): necessary and sufficient for A abelian is that G be a nil-2 group, with further restrictions in the Z_p^n case.","tokens_in":46826,"tokens_out":15570,"duration_ms":96380,"significance":"If the results hold, Theorem IV.8 is a clean and genuinely useful classification: it extends the Dehaene-De Moor and Hostens et al. characterizations of Pauli stabilizer states (Refs. [56,48]) to every level of the Clifford hierarchy, and pins the wavefunction-phase level exactly one above the stabilizer level, matching the authors' k=1 and p=2 anchors. The manuscript's strengths are real: full proofs are supplied (App. C), including the finite-difference characterization of the diagonal Clifford hierarchy (Thm. A.8) and the affine-composition invariance theorem (Thm. A.11); the classification is constructive (Algorithm 1); Lemma V.3's level-2 realization of type-III TQDs, the carry-function degree computation (degree-p exponent for type-I/II), and the deterministic qudit magic-state protocol are all proven; and Appendix D's nil-2 characterization of D(G)=D^ω(A) is a substantive standalone result. The conjectural parts are honestly labeled and falsifiable: the k=p lower bound for type-I/II cocycles is explicitly conditional on an unverified generalization of Haah's Wilson-line method (Sec. V, b386) and does not bear on the proven classification.","major_comments":[{"comment":"The k=p lower bound for the type-I/II case has no proof and rests on the explicit assumption written at b386, namely that Haah's Wilson-line extraction of topological spin from products of on-site stabilizers, proven for translation-invariant Pauli stabilizer codes (Ref. [26]), carries over to twisted quantum doubles. Consequently the paper's headline 'surprising consequence' — non-abelian type-III phases at k=2 below abelian type-I/II phases at k=p — is entirely contingent on that conjecture, and if the generalization fails the p^2-root argument collapses. The authors handle this honestly: the statements are labeled conjecture in the abstract and main text, and b370-b371 state explicitly that no lower-bound proof exists, so this is not a blocking flaw. I recommend adding one sentence in the Sec. VII discussion reiterating that the type-I/II minimum-k claim is an unproven conjecture depending on the Haah-generalization assumption, so that readers who skip Sec. V do not mistake the abelian/non-abelian level tradeoff for an established result.","section":"Sec. V, Conjecture V.4 (and Sec. I, Conjecture I.2, b73)","authors_acknowledge":"Now assume that the same procedure generalizes to TQDs.","authors_acknowledge_p":0.83}],"minor_comments":[{"comment":"Ref. [58] (Potter & Vasseur, PRB 94, 224206) is cited for the statement that a topological order with non-abelian anyons cannot be realized in a Pauli stabilizer model; that reference concerns symmetry constraints on many-body localization and does not appear to contain this theorem. For the translation-invariant models treated here the statement follows from Haah's classification (Ref. [26]); please replace or supplement the citation and state the translation-invariance hypothesis explicitly.","section":"Sec. V (type-III minimality argument)","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The claim that the Clifford hierarchy C^(k) is a group if and only if k=1 or 2 is attributed to App. A, but App. A (Lemma A.1) proves only that C^(2) is a group; the non-group statement for k ≥ 3 needs a citation (Refs. [54] or [55]) or a short explicit counterexample.","section":"Sec. III.A and Sec. I","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"In the proof of Lemma V.3 the level assignment is made for the diagonal factor Λ^ω_g of the stabilized operator A^ω_g(v) = (⊗X)Λ^ω_g; the statement of the lemma would be clearer if it said explicitly that the X part is Pauli (level 1), so the stabilizer level is set by Λ^ω_g, rather than leaving the reader to infer this from Eq. (70).","section":"Sec. V, Lemma V.3","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"Several displayed equations in Sec. IV contain a stray '/x31' glyph where the tensor-product symbol should appear; please check the final typesetting of Theorem IV.8 and Algorithm 1's output specification.","section":"Sec. IV.A, Eqs. (47) and (55)","authors_acknowledge":null,"authors_acknowledge_p":null}],"recommendation":"accept","confidential_remarks":"The proven core — Theorem IV.8, Lemma V.3's type-III level-2 realization, the Appendix D nil-2 characterization, and the deterministic qudit magic-state protocol — is coherent, novel, and consistent with the k=1 and p=2 anchors; I found the carry-function degree computation and the finite-difference arguments correct. The conjectural parts are transparently labeled, and my one substantive caveat (the conjecture-dependent 'surprising consequence') is acknowledged by the authors themselves. Two citation issues are worth an editorial check: the apparent miscitation of Ref. [58] for the non-abelian/Pauli exclusion, and the heavy reliance on very recent (2026) preprints [34-37] for the claim that nearly all Clifford-hierarchy stabilizer codes in the literature are of the XC^(k)_d form. Neither affects the mathematical content, and both can be fixed in proofs."},"author_rebuttal":null,"desk_editor":{"model":"deepseek/deepseek-v4-flash","letter":"Two things to know up front. First, the classification theorem is the paper and it is solid: Thm IV.8 is proved in full, showing XC(k)_d stabilizer states are uniform superpositions over affine-linear subspaces with a diagonal phase at level k+1 of the Clifford hierarchy, exactly one level above the stabilizer level. Second, the flashier claim — type-I/II cocycles forcing level-p stabilizers, so that non-abelian type-III phases are cheaper to realize than abelian ones — is a labeled conjecture that rests on an assumed generalization of Haah's Wilson-line argument; the authors flag the assumption explicitly at [b386] and the limitation at [b370]-[b371]. Both statements are true to the text.\n\nCredit where due. This genuinely generalizes the Dehaene-De Moor result (k=1) and the qubit XS/XP results to all k on prime qudits. The proofs are complete and coherent: normality of D_H, the coset decomposition, the affine-support argument via Lemmas IV.6/IV.7, and the finite-difference step connecting D[f] in C(k) to D[phi] in C(k+1) through Thm A.11 all check out. The carry-function degree computation behind Lemma V.3 is correct — the finite-difference argument pins degree p for type-I/II and quadratic for type-III. The type-III level-2 realization is derived, not assumed, and the k=2 lower bound there is sound because Pauli stabilizers cannot host non-abelian anyons. The magic-state counting works: N_m = 2p-1 or p-1 is never a power of p for odd p, so |CZ_m> is genuinely magic and the protocol is deterministic. The agreements with the k=1 and p=2 anchors are the right consistency checks and the paper passes them.\n\nSoft spots, in proportion. The type-I/II lower bound is the weakest step and it is exactly as conjectural as the authors say; the 'surprising consequence' in the abstract hangs entirely on it. If the Wilson-line generalization fails, that part collapses, but nothing in the classification or the type-III realization does. The generalized Bravyi-Koenig bound (Conj VII.1) is also conjecture, honestly marked. One small thing a referee should check: the citation for 'no non-abelian anyons in Pauli stabilizer models' is a Potter-Vasseur MBL paper; the standard anchor would be Haah's classification [26], and that citation warrants verification.\n\nBottom line: this deserves a serious referee and should go to review. The proven core is a real contribution with complete proofs, and the conjectural parts are cleanly separated and flagged. I would accept, with the conjectures kept labeled as they already are and that one citation checked.","headline":"The classification theorem (Thm IV.8) is proven in full and is the real result, while the type-I/II level-p claim and the \"surprising consequence\" that rides on it are honest, explicitly flagged conjectures that hinge on an unproven generalization of Haah's Wilson-line argument.","tokens_in":47427,"tokens_out":6058,"would_cite":true,"duration_ms":36443,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"pith_extraction":{"msc":["81P70","81P68"],"pacs":["03.67.Pp","03.65.Vf"],"model":"deepseek/deepseek-v4-flash","headline":"The paper proves Theorem IV.8, classifying all $XC^{(k)}_d$ stabilizer states as uniform superpositions over affine-linear subspaces with diagonal phases exactly one Clifford level above the stabilizers, and applies this to twisted…","keywords":["Clifford hierarchy","stabilizer codes","twisted quantum doubles","topological order","non-abelian anyons","qudit magic states","code switching","beyond-Pauli stabilizers"],"falsifier":"Attempt to realize the twisted quantum double $D^\\omega(\\mathbb{Z}_3^2)$ for a cocycle $\\omega$ with a type-I or type-II component using an $XC^{(2)}_d$ code with Clifford ($k=2$) stabilizers. A construction whose unique codestate lies in that phase would refute the conjecture, whereas proving that every $XC^{(k)}_d$ stabilizer state with $k<p$ can only exhibit $p$-th roots of unity as topological spins would confirm the $k=p$ bound.","tokens_in":47915,"feed_emoji":"⚛️","tokens_out":9713,"duration_ms":48175,"temperature":0.7,"pith_summary":"This paper introduces the $XC^{(k)}_d$ stabilizer formalism, in which codes are built from products of qudit Pauli-$X$ operators and diagonal gates drawn from a fixed level $k$ of the Clifford hierarchy on prime-dimensional qudits, and proves a complete classification of their unique code states. The central result, Theorem IV.8, states that any such state is, up to permutation of qudits, a uniform superposition over an affine-linear subspace of $\\mathbb{Z}_p^n$ carrying a diagonal phase from level $k+1$ of the hierarchy, exactly one level above the stabilizers themselves. The paper applies this formalism to twisted quantum doubles, proving that every 2D twisted quantum double $D^\\omega(\\mathbb{Z}_p^n)$ with $\\omega$ a product of type-III cocycles and $p$ an odd prime is realizable with Clifford ($k=2$) stabilizers. It conjectures that cocycles containing type-I or type-II components force level-$p$ stabilizers for $p$-dimensional qudits, and shows how measuring $XC^{(k)}_d$ stabilizers prepares a logical magic state in a pair of qudit toric codes deterministically.","feed_headline":"New stabilizer theorem pins phases one Clifford level higher","feed_subtitle":"Level-k stabilizers carry level-k+1 phases; non-abelian twisted doubles need only Clifford gates.","key_machinery":"The load-bearing object is the diagonal subgroup of the Clifford hierarchy, $C^{(k),n}_d$, characterized in Theorem A.8 through finite difference operators: a diagonal gate $D[f]$ lies at level $k$ exactly when $k+1$ successive finite differences of its phase function $f$ vanish, and by Corollary A.9 each finite difference lowers the level by one. Two structural facts carry the classification: the diagonal subgroup $D_H$ of a stabilizer group $H$ is normal, so each stabilizer factorizes uniquely as a Pauli-$X$ string times a diagonal gate; and Lemma IV.3 shows a computational-basis vector belongs to the state's support precisely when every diagonal stabilizer evaluates to $+1$ on it. These force the support to be an affine-linear subspace (Corollary IV.7) and pin the phase at one level above the stabilizers via Theorem A.11, which states that affine-linear pullback preserves Clifford-hierarchy level.","core_discovery":"The paper's principal claim is Theorem IV.8: a state is an $XC^{(k)}_d$ stabilizer state if and only if, after a permutation of qudits, it equals $$\\frac{1}{$p^{{\\ell/2}}$}\\sum_{x\\in\\mathbb{Z}_p^\\ell} D[\\varphi]\\,|x, Mx+c\\rangle,$$ where $M$ is an $(n-\\ell)\\times\\ell$ matrix over $\\mathbb{Z}_p$, $c\\in\\mathbb{Z}_p^{n-\\ell}$, and the diagonal operator $D[\\varphi]$ lies at level $k+1$ of the Clifford hierarchy. This means the wavefunction phase sits exactly one level above the stabilizer level: producing level-$k+1$ phases in a computational-basis expansion requires stabilizers at level $k$. The authors then prove that every 2D twisted quantum double $D^\\omega(\\mathbb{Z}_p^n)$ with $\\omega$ a product of type-III cocycles and $p$ an odd prime admits a realization with Clifford ($k=2$) stabilizers; because such topological orders host non-abelian anyons and Pauli stabilizer models cannot realize non-abelian orders, this lower bound is saturated. They conjecture, with supporting evidence but no proof, that any cocycle containing a type-I or type-II component forces the minimal level up to $k=p$.","pith_inferences":["The 'phase sits one level above' step is likely a general principle of monomial-style stabilizer formalisms: any code-switching protocol passing through non-Pauli stabilizer codes pays exactly one extra Clifford level in the wavefunction, which may carry over to composite-dimensional qudits and to the 3D twisted models the authors flag as amenable.","If the $k=p$ bound holds, the intuition that non-abelian topological order is 'costlier' is inverted in stabilizer terms: abelian twisted orders with type-I or type-II cocycles become the expensive ones, which may shift which intermediate codes fault-tolerant non-Clifford constructions favor.","The determinism of the qudit magic-state protocol suggests an odd-prime advantage over qubits: the measurement outcomes that make the qubit protocol probabilistic do not appear for $p\\ge 3$, a claim worth testing numerically for small $p$ and through the fault-tolerance of the switching steps.","A testable extension of the type-III construction is to check whether the $k=2$ realization generalizes to 3D twisted quantum doubles and twisted fracton models, which the authors identify as formally similar; success would extend the classification beyond 2D."],"forward_implications":["Theorem IV.8 gives a complete structural description of unique code states of beyond-Pauli stabilizer codes: uniform superpositions over affine-linear subspaces with diagonal phases one Clifford level above the stabilizers, together with an algorithm converting any $XC^{(k)}_d$ subgroup into canonical form.","Every 2D twisted quantum double $D^\\omega(\\mathbb{Z}_p^n)$ with $\\omega$ a product of type-III cocycles and $p$ an odd prime is realizable with Clifford ($k=2$) stabilizer codes, giving a saturated lower bound $k=2$ for these non-abelian phases.","If Conjectures I.2 and V.4 hold, topological orders with non-abelian anyons (type-III cocycles) are realizable at lower stabilizer level than purely abelian twisted orders (type-I or type-II cocycles), for prime qudit dimension at least 3; the two cases coincide for qubits.","The measurement protocol prepares a logical magic state in a pair of qudit toric codes deterministically for odd prime $p$, improving on the probabilistic qubit protocol it generalizes.","A conjectured generalized Bravyi-Koenig bound (Conjecture VII.1) would restrict locality-preserving logical gates on level-$k$ XC codes in $d$ dimensions to level $d+k-1$ of the Clifford hierarchy."],"supporting_citations":[{"why":"Supplies the characterization of the diagonal Clifford hierarchy $C^{(k),n}_d$ via polynomial degree and the finite-difference criterion (Eq. 14, Theorem A.8) that undergirds Theorem IV.8.","marker":"[54]"},{"why":"Supplies the Wilson-line method extracting topological spin from products of on-site stabilizers for translation-invariant Pauli stabilizer codes, whose generalization to twisted quantum doubles the $k=p$ conjecture presupposes.","marker":"[26]"},{"why":"The stabilizer-state classification over GF(2) (affine subspaces with linear and quadratic phases) that Theorem IV.8 generalizes to $XC^{(k)}_d$ states.","marker":"[56]"},{"why":"Sources the decomposition of 3-cocycles into type-I, -II, and -III contributions and the fact that type-III cocycles produce non-abelian anyons, used throughout Section V.","marker":"[50]"},{"why":"Defines the twisted quantum double Hamiltonian whose star and plaquette stabilizers are identified with an $XC^{(k)}_d$ code in Lemma V.3 and whose spin data drive Lemma D.16.","marker":"[52]"},{"why":"Provides the qubit code-switching and magic-state preparation protocol that Section VI generalizes to odd prime qudits, with the qudit version made deterministic.","marker":"[34]"},{"why":"Establishes that non-abelian topological order cannot be realized by Pauli stabilizer models, giving the $k=1$ impossibility that makes $k=2$ saturated for type-III cocycles.","marker":"[58]"}],"fun_headline_variants":["Level-k stabilizers host level-k+1 phases","Twisted quantum doubles via Clifford stabilizers","Stabilizer framework lifts phases one level up","Clifford stabilizers realize non-abelian doubles","Beyond-Pauli stabilizers for twisted doubles"],"cache_read_input_tokens":0,"weakest_assumption_plain":"The conjectured lower bound $k=p$ for type-I and type-II cocycles rests on the unproven premise that the Wilson-line method of reading topological spin off products of on-site stabilizers, established for translation-invariant Pauli stabilizer codes in a cited source, carries over unchanged to twisted quantum doubles; the authors flag this step by writing 'now assume that the same procedure generalizes to TQDs' and acknowledge that no lower-bound proof exists.","fun_headline_variants_meta":{"raw":{"variants":["Level-k stabilizers host level-k+1 phases","Twisted quantum doubles via Clifford stabilizers","Stabilizer framework lifts phases one level up","Clifford stabilizers realize non-abelian doubles","Beyond-Pauli stabilizers for twisted doubles"]},"model":"deepseek/deepseek-v4-flash","effort":"low","cost_usd":0.000163,"raw_usage":{"cost":0.000162708,"is_byok":false,"cost_details":{"upstream_inference_cost":0.000162708,"upstream_inference_prompt_cost":0.000134428,"upstream_inference_completions_cost":2.828e-05},"total_tokens":1266,"prompt_tokens":1165,"completion_tokens":101,"prompt_tokens_details":{"audio_tokens":0,"video_tokens":0,"cached_tokens":256,"cache_write_tokens":0},"completion_tokens_details":{"audio_tokens":0,"image_tokens":0,"reasoning_tokens":27}},"tokens_in":1165,"tokens_out":101,"cost_source":"openrouter_usage","duration_ms":1657,"image_pages":[],"temperature":1.0,"billed_attempts":1,"reasoning_tokens":27,"transport_provider":"openrouter","cache_read_input_tokens":0,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-10-03T03:17:13.731473+00:00","model_set":{"reader":"deepseek/deepseek-v4-flash"},"falsifier":"Attempt to realize the twisted quantum double $D^\\omega(\\mathbb{Z}_3^2)$ for a cocycle $\\omega$ with a type-I or type-II component using an $XC^{(2)}_d$ code with Clifford ($k=2$) stabilizers. A construction whose unique codestate lies in that phase would refute the conjecture, whereas proving that every $XC^{(k)}_d$ stabilizer state with $k<p$ can only exhibit $p$-th roots of unity as topological spins would confirm the $k=p$ bound.","supporting_citations":[{"cited_title":"Diagonal gates in the Clifford hierarchy,","cited_arxiv_id":null,"evidence_quote":"Supplies the characterization of the diagonal Clifford hierarchy $C^{(k),n}_d$ via polynomial degree and the finite-difference criterion (Eq. 14, Theorem A.8) that undergirds Theorem IV.8."},{"cited_title":"Classification of translation invariant topolog- ical Pauli stabilizer codes for prime dimensional qudits on two-dimensional lattices,","cited_arxiv_id":null,"evidence_quote":"Supplies the Wilson-line method extracting topological spin from products of on-site stabilizers for translation-invariant Pauli stabilizer codes, whose generalization to twisted quantum doubles the $k=p$ conjecture presupposes."},{"cited_title":"Clifford group, stabilizer states, and linear and quadratic operations over GF(2),","cited_arxiv_id":null,"evidence_quote":"The stabilizer-state classification over GF(2) (affine subspaces with linear and quadratic phases) that Theorem IV.8 generalizes to $XC^{(k)}_d$ states."},{"cited_title":"Topological interactions in broken gauge theories","cited_arxiv_id":"hep-th/9511195","evidence_quote":"Sources the decomposition of 3-cocycles into type-I, -II, and -III contributions and the fact that type-III cocycles produce non-abelian anyons, used throughout Section V."},{"cited_title":"Twisted Quantum Double Model of Topological Phases in Two--Dimension","cited_arxiv_id":"1211.3695","evidence_quote":"Defines the twisted quantum double Hamiltonian whose star and plaquette stabilizers are identified with an $XC^{(k)}_d$ code in Lemma V.3 and whose spin data drive Lemma D.16."},{"cited_title":"Uni- versal fault-tolerant quantum computation in 2d with- out getting tied in knots,","cited_arxiv_id":null,"evidence_quote":"Provides the qubit code-switching and magic-state preparation protocol that Section VI generalizes to odd prime qudits, with the qudit version made deterministic."},{"cited_title":"Symmetry constraints on many-body localization,","cited_arxiv_id":null,"evidence_quote":"Establishes that non-abelian topological order cannot be realized by Pauli stabilizer models, giving the $k=1$ impossibility that makes $k=2$ saturated for type-III cocycles."}],"review_version":1}