{"id":"64fbfa33-eac7-4a08-b021-e145fc8d9df7","arxiv_id":"1908.02599","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Any topological order in a non-Abelian family can be described by a vector of Abelian anyons and one symmetric matrix K, generalizing the Abelian K matrix formalism.","lead":"This paper gives a compact K matrix description for topological orders that share the same non-Abelian anyon content, where the matrix is corrected by the mutual statistics of Abelian anyons. It promises a fast way to generate the data of many 2+1D topological phases, which matters for classifying materials and for quantum computation.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Theorem 1 asserts rank |det K| N_C without proving integrality; rational K satisfying the stated hypotheses can yield non-integer ranks for valid C, so the matrix description is not yet shown to always produce a topological order.","rationale":"The reader's weakest assumption concerns the completeness of the generalized hierarchy construction imported from Ref. 6, which is external to this paper. I identified a different, more internal gap: Theorem 1's rank formula is asserted without proving that |det K| N_C is an integer for every allowed (a,K). Because K can be rational while satisfying the integrality conditions, the claimed rank can be a rational non-integer unless a hidden divisibility relation N_C * 2s_a in Z holds for all Abelian anyons in a modular category. The paper's induction does not establish this relation; it merely multiplies determinants. This is a concrete, checkable condition, and if it fails the matrix construction is invalid for a valid input. The concern does not change the overall verdict: the paper is plausible and the gap is likely repairable by adding an explicit integrality hypothesis or proving the divisibility lemma, so CONDITIONAL remains appropriate. Since the reader's verdict is already CONDITIONAL and my concern reinforces that status rather than overturning it, I keep the reader's verdict unchanged.","tokens_in":8792,"tokens_out":38829,"duration_ms":424654,"concrete_test":"Using the Rowell-Stong-Wang classification of modular categories of small rank or a Sage MTC library, enumerate all modular categories of rank at most 6. For each Abelian anyon a, compute N_C * 2s_a. If any pair yields a non-integer (in particular, any category with an order-3 Abelian anyon and rank not divisible by 3), evaluate the one-step K = -2s_a (or K = 2 - 2s_a) and check whether |det K| N_C is an integer. A non-integer value is a direct counterexample to Theorem 1's rank formula; if none is found, the missing divisibility lemma should be stated and proved in a revision.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Theorem 1's rank formula R = |det K| N_C is load-bearing. The hypotheses (det K != 0; K_IJ + t_{a_I,a_J} in Z; K_II + t_{a_I,a_I} in 2Z) do not imply |det K| N_C is an integer, since K is a rational symmetric matrix. For one step, K_11 = m_c - 2s_a with m_c even. If C contains an Abelian anyon a of order n with 2s_a = r/n, then K_11 satisfies the stated integrality conditions, but R = |m_c - 2s_a| N_C can be fractional whenever n does not divide N_C. The induction in Appendix A computes det K_1 = det K_0 (m_c - 2s_a) and identifies this with the rank increment, but it never proves the resulting product is an integer. If such a pair (C,a) exists - for example, a rank-4 modular category with a Z_3 Abelian anyon - the theorem assigns C_{a,K} a non-integer number of anyon types, so the construction cannot describe a topological order. The paper should either prove the hidden divisibility lemma N_C * 2s_a in Z for all Abelian anyons a in a modular category, or add |det K| N_C in Z as an explicit hypothesis.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proposes a K-matrix-style parametrization for non-Abelian families of 2+1D topological orders. Given a starting topological order C, any topological order in the same non-Abelian family (defined via the generalized hierarchy construction of Ref. 6) is claimed to be described by a tuple a=(a_I) of Abelian anyons in C together with a symmetric invertible matrix K with K_IJ = k_IJ - t_{a_I,a_J}, where K_IJ + t_{a_I,a_J} are integers and K_II + t_{a_I,a_I} are even integers. Theorem 1 gives explicit formulas for the fusion rules, topological spins, S matrix, rank, and chiral central charge of the resulting order C_{a,K}, and the paper states that when C is a root (i.e. C_Ab is a symmetric fusion category), K becomes an integer matrix. The proof in Appendix A proceeds by induction from the one-step hierarchy construction, and a separate formal categorical formulation is sketched. The paper also discusses equivalence relations on (a,K) and formulates a conjecture characterizing when two such descriptions give equivalent topological orders.","tokens_in":9071,"tokens_out":21007,"duration_ms":242899,"significance":"If the central claim holds, the paper provides a compact, algorithmically usable description of potentially infinite families of topological orders from a single root category, which is a useful organizing principle for classification and for generating explicit data. The linear-algebra formulas for fusion, spin, S, rank, and central charge are concrete and directly implementable. The inductive proof in Appendix A is a genuine verification of the structural formulas, and the determinant and chiral central charge formulas are explicit. The main weakness, as discussed below, is that a load-bearing integrality statement is asserted without proof, and a few derivations are only sketched; these issues are fixable but need to be addressed before the theorem is fully rigorous.","major_comments":[{"comment":"The theorem's hypotheses allow K to be a rational symmetric matrix, since t_{a_I,a_J} are rational in general. The rank formula N_{C_{a,K}} = |det K| N_C therefore requires a proof that |det K| N_C is an integer. The induction in Appendix A only computes det K_1 = (m_c - 2s_{a_c}) det K_0 and identifies this with the rank increment; it never establishes that the resulting product is an integer. For a one-step construction, this reduces to the divisibility condition N_C * 2s_a ∈ Z for every Abelian anyon a in every category that arises in the hierarchy, which is not stated or proved. The paper should either provide a proof or a precise citation for this divisibility lemma, or add the explicit hypothesis |det K| N_C ∈ Z to Theorem 1.","section":"Theorem 1 (rank bullet) and Appendix A, Eq. (A5)"},{"comment":"The claim that the assumption det K_0 ≠ 0 is inessential and can be dropped is supported only by the statement of the GL(Z) equivalence C_{a,K} ≃ C_{W a, W K W^T}. No argument is given that for an arbitrary K satisfying the hypotheses one can choose W ∈ GL(κ,Z) so that all leading principal minors of W K W^T are nonzero while preserving the integrality conditions. This is a standard genericity fact, but it should be either proved or replaced by a precise reference.","section":"Appendix A, paragraph 'In the above proof...'"},{"comment":"The S matrix formula is asserted to 'follow directly' from the verified fusion, spin, and equivalence relations, but no derivation is included. Since the S matrix is a central part of Theorem 1 and its normalization with |det K| and the phase factor are nontrivial for rational K, a sketch of the derivation (or a reference to a full proof) should be supplied.","section":"Eq. (17) and Appendix A"}],"minor_comments":[{"comment":"The sentence 'We showed that given a topological order C, any topological order in the same non-Abelian family can be efficiently represented...' states the paper's main claim, but the formal theorem in the text only describes the result of a finite sequence of hierarchy steps; the surjectivity statement should be stated explicitly as an assumption or with a reference to Ref. 6.","section":"Introduction"},{"comment":"The data of the one-step hierarchy construction, including the rank formula |M_c| N_C, are listed without derivation; a citation to Ref. 6 at that point would clarify which results are being imported.","section":"One-step construction, Eqs. (9)-(12)"},{"comment":"The grading group Z^κ/K(2 ker a,−) is written as a quotient of the dual space by a sublattice, which as written is an infinite group; the intended finite quotient  appears only after imposing the condition f(-)+t(i,a(-)) ∈ Z. A clarifying sentence would prevent confusion.","section":"Formal categorical formulation"},{"comment":"The word 'Furture' should be 'Future'.","section":"Conclusion"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is a compact letter that builds heavily on the author's own prior hierarchy construction (Ref. 6) and thesis (Ref. 11). The main concern is completeness of the proof rather than correctness: the integrality of the rank formula should be resolved with a short lemma or an added hypothesis, and the S-matrix derivation should be supplied. If the journal's format permits a quickly added appendix, these are manageable revisions. The paper is likely of interest to the topological order community, but in its current form the central theorem is not fully self-contained."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Here's my take on arXiv:1908.02599. The genuinely new thing is a closed-form K-matrix description for non-Abelian families: instead of iterating the hierarchy construction step by step, one writes a single symmetric matrix K_IJ = k_IJ - t_{a_I,a_J} with a_I Abelian anyons in a seed category C. For root C the K matrix is integer. That is a substantial upgrade over Ref. 6, and the categorical formulation with grading and condensation is original, not a relabeling. The paper does a good job of making the construction concrete: Theorem 1 gives explicit fusion, spin, and rank formulas, and the induction in Appendix A checks the key steps. I believe the determinant and chiral central charge formulas are right.\n\nNow the soft spots. The S-matrix formula is stated without a real derivation; the induction proof says it 'follows directly' but doesn't show it. The removal of the det K0 != 0 assumption is sketched via GL(Z) equivalence and is plausible but not fully worked out. Both are minor for a letter, but a serious referee should ask for the S-matrix argument.\n\nThe stress-test note about non-integer rank does not hold up as stated. A rank-4 modular category with a Z_3 Abelian anyon cannot exist: the subgroup generated by an Abelian anyon acts freely on the simple objects, so the total rank is divisible by the order of the anyon. The paper never states that divisibility, and the proof of R = |det K| N_C does not show the product is integral, so there is a real gap in rigor. But the categorical construction defines a fusion category at the end, so the rank is an integer by construction; the formula is almost certainly correct and needs a small lemma, not a fundamental fix.\n\nThe paper builds on the author's own prior hierarchy classification, and the root characterization is self-cited, but that's appropriate here since the prior work is the established foundation. No fitting or circular reasoning.\n\nBottom line: this is a useful tool for anyone classifying topological orders or generating anyon models. It deserves peer review. I'd send it to a competent referee and ask for a tightened proof of the rank integrality and the S matrix. With those, it should be publishable.","headline":"A genuinely useful matrix formulation for non-Abelian families, with solid but terse proofs; the rank-integrality gap is patchable, not fatal.","tokens_in":9563,"tokens_out":11567,"would_cite":true,"duration_ms":119069,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Given a topological order $\\mathcal C$, the paper proves that every topological order in its non-Abelian family is described by a tuple of Abelian anyons and a symmetric invertible $K$ matrix, with closed formulas for the fusion, spin…","keywords":["K matrix formulation","non-Abelian anyons","2+1D topological orders","non-Abelian family","hierarchy construction","Laughlin states","modular tensor categories","anyon condensation"],"falsifier":"Compute the complete fusion and modular data of a topological order that is claimed to belong to a non-Abelian family whose root has an Abelian anyon with non-integral self-statistics; if no pair $(\\boldsymbol a,K)$ over a symmetric root reproduces that data, the matrix formulation misses members of the family or the root characterization is wrong.","tokens_in":8592,"feed_emoji":"🧮","tokens_out":11784,"duration_ms":99030,"temperature":0.7,"pith_summary":"This paper claims that every topological order in the same non-Abelian family as a given topological order $\\mathcal C$ can be written compactly as $\\mathcal C_{\\boldsymbol a,K}$: choose a tuple $\\boldsymbol a=(a_I)$ of Abelian anyons inside $\\mathcal C$ and a symmetric invertible matrix $K$ with entries $K_{IJ}=k_{IJ}-t_{a_I,a_J}$, where $k_{IJ}$ are integers with even diagonal and $t_{a_I,a_J}$ are the mutual statistics between the chosen anyons. The claim matters because the previous hierarchy construction had to be applied step by step, making it hard to reach a relative that is many steps away; the matrix description turns any member of the family into data that can be written down at once. Theorem 1 gives explicit formulas for the fusion rules, topological spins, modular $S$ matrix, rank, and chiral central charge of $\\mathcal C_{\\boldsymbol a,K}$ in terms of $K$ and the mutual statistics. When $\\mathcal C$ is a root of the family, whose Abelian anyons are mutually trivial bosons or fermions, $K$ becomes an integer matrix, so generating large numbers of topological orders becomes a matter of enumerating integer matrices.","feed_headline":"One K matrix describes whole non-Abelian families","feed_subtitle":"One matrix encodes fusion, spin, S matrix, rank, and central charge for every member of a non-Abelian family.","key_machinery":"The central object is the generalized $K$ matrix, $K_{IJ}=k_{IJ}-t_{a_I,a_J}$: an integer symmetric part $k_{IJ}$ (even on the diagonal) corrected by subtracting the mutual statistics $t_{a_I,a_J}$ of the Abelian anyons $a_I,a_J$ chosen in $\\mathcal C$. It enters as the exponent of the multilayer Laughlin wavefunction and it encodes, in one matrix, all the information that the step-by-step hierarchy construction would otherwise take many steps to accumulate. Theorem 1 uses this single matrix to write the equivalence relations on anyon labels, the fusion rule, topological spin, $S$ matrix, rank, and chiral central charge of the constructed topological order. The integrality of $K$ for roots is what reduces the enumeration of large numbers of topological orders to integer linear algebra.","core_discovery":"The paper's central claim is a matrix formulation for non-Abelian families of 2+1D topological orders. Starting from a topological order $\\mathcal C$, the paper constructs a new topological order by letting Abelian anyons $a_I$ of $\\mathcal C$ form a multilayer Laughlin-like state, and summarizes the construction by a pair $(\\boldsymbol a,K)$ with $K_{IJ}=k_{IJ}-t_{a_I,a_J}$. The anyons of the resulting order are pairs $(i,\\boldsymbol l)$ with $i$ an anyon of $\\mathcal C$ and $\\boldsymbol l$ an integer vector, modulo the equivalence relation $(i,\\boldsymbol l)\\sim(i\\otimes a_I,\\boldsymbol l+K_I-\\boldsymbol t_i+\\boldsymbol t_{i\\otimes a_I})$. Theorem 1 states the resulting fusion rule $(i,\\boldsymbol l)\\otimes(j,\\boldsymbol k)=\\oplus_s N^s_{ij}(s,\\boldsymbol l+\\boldsymbol k-\\boldsymbol t_i-\\boldsymbol t_j+\\boldsymbol t_s)$, the topological spin $s_{(i,\\boldsymbol l)}=s_i+\\frac{1}{2}(\\boldsymbol l-\\boldsymbol t_i)^T K^{-1}(\\boldsymbol l-\\boldsymbol t_i)$, the $S$-matrix formula $S_{(i,\\boldsymbol l),(j,\\boldsymbol k)}=|\\det K|^{-1/2}S_{ij}e^{-2\\pi i(\\boldsymbol l-\\boldsymbol t_i)^T K^{-1}(\\boldsymbol k-\\boldsymbol t_j)}$, the rank $|\\det K|N_{\\mathcal C}$, and the chiral central charge $c_{\\mathcal C}+\\operatorname{sgn}K$. For a root $\\mathcal C$, the Abelian sector is a symmetric fusion category and $K$ is an integer matrix, with the parity of $K_{II}$ indicating whether $a_I$ is a boson or a fermion. The paper also gives a basis-independent categorical construction and a conjecture characterizing when two pairs $(\\boldsymbol a,K)$ produce equivalent topological orders.","pith_inferences":["Editorial inference: if Conjecture 1 is correct, deciding whether two $K$ matrices describe the same phase becomes a well-defined reduction problem under $\\operatorname{GL}(\\mathbb Z)$ transformations and the addition or removal of trivial bilayers, which could be turned into an algorithmic check.","Editorial inference: the multilayer-Laughlin assumption also suggests a sharp boundary for the classification: if non-Abelian anyons can form nontrivial collective states other than Laughlin states, the non-Abelian family equivalence would need additional moves beyond adding Abelian anyons.","Editorial inference: because the root plus $K$ determines the anyon data, quantities relevant to topological quantum computation, such as the set of braiding phases, are functions of the root and $K$; comparing these functions across a family could show which non-Abelian properties are family invariants."],"forward_implications":["For any root $\\mathcal C$, every topological order in its non-Abelian family is obtained from some pair $(\\boldsymbol a,K)$; no sequence of intermediate hierarchy steps needs to be tracked.","The physical data of each generated order—fusion multiplicities, topological spins, modular $S$ matrix, rank, and chiral central charge—are given by closed formulas in terms of $K$ and the mutual statistics of the base order.","Because roots have integer $K$, enumerating candidate topological orders by determinant and rank reduces to enumerating integer symmetric matrices satisfying the parity constraints.","Known topological orders can be grouped into non-Abelian families by identifying the root and the pair $(\\boldsymbol a,K)$ that produces them; unknown orders can be generated from the same root.","The classification problem for 2+1D topological orders is reduced to classifying roots, since every other phase in a family is a $K$-matrix construction over its root."],"supporting_citations":[{"why":"defines the generalized hierarchy construction and the non-Abelian family equivalence that this paper reformulates as matrices.","marker":"[6]"},{"why":"gives the Abelian K-matrix formulation that the non-Abelian generalization reduces to when the base order is trivial.","marker":"[7]"},{"why":"supplies the Laughlin wavefunction that Abelian anyons are made to form in each hierarchy step.","marker":"[8]"},{"why":"characterizes roots as orders whose Abelian anyons form a symmetric fusion category, making K integral.","marker":"[11]"},{"why":"provides the anyon-condensation operation used to remove the symmetric subcategory in the categorical formulation.","marker":"[13]"}],"fun_headline_variants":["K matrix captures whole non-Abelian families","One K matrix encodes every family member","Unified matrix for non-Abelian topological orders","Single K matrix tames non-Abelian families","Matrix formulation for all non-Abelian orders"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that repeatedly letting Abelian anyons condense into Laughlin-like states can reach every topological order in a non-Abelian family, and that every family has a smallest 'root' order in which all Abelian anyons are mutually trivial bosons or fermions.","fun_headline_variants_meta":{"raw":{"variants":["K matrix captures whole non-Abelian families","One K matrix encodes every family member","Unified matrix for non-Abelian topological orders","Single K matrix tames non-Abelian families","Matrix formulation for all non-Abelian orders"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000801,"raw_usage":{"total_tokens":3601,"prompt_tokens":1107,"completion_tokens":2494,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":723,"completion_tokens_details":{"reasoning_tokens":2423}},"tokens_in":723,"tokens_out":2494,"duration_ms":19617,"temperature":1.0,"reasoning_tokens":2423,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T14:39:26.335623+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the complete fusion and modular data of a topological order that is claimed to belong to a non-Abelian family whose root has an Abelian anyon with non-integral self-statistics; if no pair $(\\boldsymbol a,K)$ over a symmetric root reproduces that data, the matrix formulation misses members of the family or the root characterization is wrong.","supporting_citations":[{"cited_title":"Hierarchy construction and non-Abelian families of generic topological orders","cited_arxiv_id":"1701.07820","evidence_quote":"defines the generalized hierarchy construction and the non-Abelian family equivalence that this paper reformulates as matrices."},{"cited_title":"Lan ,\\ title A Classification of (2+1)D Topological Phases with Symmetries ,\\ https://uwspace.uwaterloo.ca/handle/10012/12389 Ph.D","cited_arxiv_id":null,"evidence_quote":"characterizes roots as orders whose Abelian anyons form a symmetric fusion category, making K integral."}],"review_version":1}