{"id":"a257d2b4-6198-41b5-969f-24c91d5da6e8","arxiv_id":"2607.07227","paper_version":1,"verdict":"ACCEPT","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"unknown","formal_verification":"none","parameter_count":0,"one_line_summary":"Katsura algebra simplicity is completely characterized by explicit matrix conditions (Theorems 5.7, 6.14) and shown to be decidable in polynomial space.","lead":"The paper gives a complete characterization of when Katsura algebras and their faithful quotients are simple, in terms of explicit conditions on two integer matrices A and B. It also provides polynomial-space algorithms to decide simplicity, making this the first decidable non-contracting self-similar groupoid family in the non-Hausdorff setting.","discovery_kind":"unclear","skeptic_critique":{"model":"glm-5.2","headline":"No significant objection identified","rationale":"The reader correctly identifies the dependency on [8] and [14] as the most external, least self-contained link in the argument. However, upon examination, the specific conditions [8] requires (amenability and abelian isotropies of G^0(G,E)) are explicitly verified in Lemma 3.1 via a direct computation, and the amenability transfer from G(G,E) to G^0(G,E) via [14] is a standard property of open subgroupoids of amenable groupoids. The concern is legitimate as a dependency on recent preprints, but it does not rise to a load-bearing objection against the paper's internal logic: the paper does the necessary work to satisfy the hypotheses of the cited results. The combinatorial characterizations (Theorems 5.7, 6.14) and the algorithmic results (Section 7) follow from established inverse semigroup algebra theory (Propositions 2.3-2.6) and careful case analysis. The examples in Section 8 demonstrate independence of the two singular ideal conditions. No internal inconsistency or gap was found in the central argument. The verdict of ACCEPT at MODERATE confidence is appropriate; the moderate confidence reflects the dependency on recent results [8, 9], which is an honest assessment rather than a flaw in the paper itself.","tokens_in":28010,"tokens_out":934,"duration_ms":554777,"concrete_test":"Independently verify the commutativity computation in Lemma 3.1: take two arbitrary elements [qgq*,p] and [qrhr*q*,p] in the isotropy of G^0(G,E) at p, and confirm that their product equals [qrhr*q*,p][qgq*,p] using only the relation g(r)g|r = rg|r (from the self-similar action) and the commutativity of (G)_{s(r)}. If the four-step computation in the proof does not close, the abelianity claim fails and Theorem 3.3 cannot invoke [8].","verdict_should_be":"UNCHANGED","load_bearing_attack":"The reader's weakest_assumption concerns the amenability transfer from G(G,E) to G^0(G,E) via [14, Prop. 2.17] and [14, Thm. 2.18], and the application of [8, Thm. D and Thm. G] to G^0(G,E). I examined this chain carefully. Lemma 3.1 establishes that G^0(G,E) is amenable with abelian isotropies. The amenability argument is standard: G is a bundle of abelian (hence amenable) groups, so G is amenable; G(G,E) is amenable by [14, Thm. 2.18]; G^0(G,E) is an open subgroupoid of an amenable groupoid, hence amenable by [14, Prop. 2.17(i)]. The abelianity of isotropies of G^0(G,E) is verified directly in Lemma 3.1 by an explicit computation showing that any two elements [qgq*,p] and [qrh r*q*,p] of the isotropy at p commute, using the commutativity of the underlying group bundle. This is correct. The application of [8] requires amenability and abelian isotropies of G^0(G,E), both of which are established. The faithful conditional expectation E: C*_r(G(G,E)) -> C*_r(G^0(G,E)) from [7, Lemma 3.5] with E(J) = J^0 (Lemma 3.2) is a standard tool. The overall logic of Theorem 3.3 is: J_C = 0 implies J^0_C = 0 (by definition); [8] gives J^0 = 0 from J^0_C = 0 (using amenability + abelian isotropies); Lemma 3.2 gives J = 0 from J^0 = 0. This chain is sound. The combinatorial characterizations in Theorems 5.7 and 6.14 follow from Propositions 2.3-2.6, which are standard results from [1, 24] about inverse semigroup algebras. The algorithmic complexity analysis in Section 7 is careful about space bounds. I do not find a load-bearing concern that would undermine the central claims.","agreement_with_reader":"partial"},"referee_report":{"model":"glm-5.2","summary":"This paper gives a complete characterization of when the singular ideal vanishes for both Katsura-Exel-Pardo (KEP) groupoid algebras and their faithful quotients, stated purely in terms of the input matrices A and B. The key results are Theorem 5.7 (non-faithful case) and Theorem 6.14 (faithful case), which combined with minimality and topological freeness conditions yield complete simplicity characterizations (Corollaries 5.8 and 6.15). A crucial intermediate result, Theorem 3.3, establishes that vanishing of the C*-singular ideal is equivalent to vanishing of the algebraic (Steinberg algebra) singular ideal, reducing the analytic question to a combinatorial one. The authors further provide polynomial-space algorithms (Section 7) to decide these conditions, and give examples (Section 8) demonstrating independence of the singular ideal conditions between the faithful and non-faithful settings.","tokens_in":28392,"tokens_out":1444,"duration_ms":413278,"significance":"The paper makes a substantial contribution to the study of non-Hausdorff groupoid C*-algebras and Steinberg algebras. Katsura algebras are central in the classification of Kirchberg algebras, and characterizing their simplicity in the non-Hausdorff case has been an open challenge, with the singular ideal being the main obstruction. The reduction from C*-simplicity to algebraic simplicity (Theorem 3.3) is a clean and useful bridge. The matrix-level characterizations in Theorems 5.7 and 6.14 are the first such complete characterizations for non-contracting self-similar groupoids. The polynomial-space algorithms are a notable strength, as they make the simplicity conditions effectively decidable. The introduction of c-stable subgroups (Definition 6.5) as a computationally tractable substitute for recurring subgroups is a valuable technical innovation. The examples in Section 8, particularly those showing independence of the singular ideal conditions, add concrete value.","major_comments":[{"comment":"Theorem 3.3 is load-bearing for the entire paper, as all subsequent characterizations of C*-simplicity rely on the equivalence between vanishing of J and vanishing of J_C. The proof applies [8, Thm. D and Thm. G] to the gauge-invariant subgroupoid G^0(G,E), which requires amenability and abelian isotropies (Lemma 3.1). The amenability transfer from G(G,E) to G^0(G,E) via [14, Prop. 2.17(i)] and [14, Thm. 2.18] appears standard, and the abelianity computation in Lemma 3.1 is correct. However, the authors should explicitly state in the proof of Theorem 3.3 (or in Lemma 3.1) which specific hypotheses of [8, Thm. D and Thm. G] are being verified, to make the logical chain fully transparent to the reader. As stated, the reader must independently check that G^0(G,E) falls within the class of groupoids covered by [8]. This is a presentation gap rather than a mathematical error, but given the重要性","section":null},{"comment":"In Algorithm 2 (line 21 of the pseudocode), the logic for checking condition (T2) appears to have a control-flow issue. The inner loop over q searches for a zero path extending p, and if found, 'continue with next p'. But if no such q is found, the code falls through to 'continue with next v' without halting and returning NO. This means the algorithm would not correctly identify a vertex v satisfying (T2). The intended logic should be: if for some p no suitable q exists, then (T2) fails for this v, so proceed to the next v; if all p have a suitable q, then (T2) holds, so halt and return NO. The current pseudocode structure does not clearly implement this. The authors should correct the control flow or clarify the intended logic.","section":null}],"minor_comments":[{"comment":"Section 2.1: The standing assumption that E has no sources is stated, but Section 2.4 explains how to handle sources by adding loops. It would help the reader to cross-reference Section 2.4 at the point where the standing assumption is first declared.","section":null},{"comment":"Definition 6.5: The term 'c-stable' is introduced, but the condition that |c: H -> H is bijective could be stated more explicitly as 'the section map h |-> h|_c restricts to a group automorphism of H'.","section":null},{"comment":"Algorithm 4: The variable names r, M, k, s, n, K are used without much explanation in the pseudocode itself. While the surrounding text explains them, adding brief inline comments would improve readability.","section":null},{"comment":"Example 8.2: The parameter b ranges over {0,1,2,3,4,5}, but the analysis groups cases as b in {0,2,3,4} and b in {1,5}. It would be clearer to state upfront which values of b lead to which outcome, perhaps in a small table.","section":null},{"comment":"Remark 8.3: The claim that 'bJK = 0, JK != 0' is exhibited by omitting v4 from Example 8.2 is somewhat terse. A brief verification that the resulting graph still satisfies (C1) would strengthen this remark.","section":null},{"comment":"Typo in Section 5, line after Proposition 5.3: 'KEP-groupoid' should be 'KEP-groupoids' for grammatical consistency.","section":null},{"comment":"Reference [8] (Gonzales-Hume) is cited as a 2026 arXiv preprint. The authors should verify if a published version is available at the time of submission.","section":null}],"recommendation":"minor_revision","confidential_remarks":"The paper is mathematically sound and the results are significant. The two major comments are both fixable: the first is a presentation clarification for Theorem 3.3, and the second is a control-flow issue in Algorithm 2's pseudocode that does not affect the theoretical correctness of Theorem 5.7. I recommend minor revision. The authors have clearly worked through substantial technical material, and the examples section demonstrates genuine understanding of the theory. The independence of the singular ideal conditions between faithful and non-faithful cases (Remark 8.3) is a nice touch that adds to the paper's contribution."},"author_rebuttal":null,"desk_editor":{"model":"glm-5.2","letter":"The main thing to know: this paper gives a complete, matrix-level characterization of when Katsura algebras and their faithful quotients are simple, including the non-Hausdorff case that was genuinely open. The conditions (T1), (T2), (C1), (C2), (FC2) are stated purely in terms of the input matrices A and B — no fitted parameters, no circularity. The polynomial-space decidability is a bonus and is, as far as I know, the first such result for non-contracting self-similar groupoids.","headline":"Complete characterization of simplicity for non-Hausdorff Katsura algebras via matrix conditions, with polynomial-space decidability","tokens_in":29195,"tokens_out":185,"would_cite":true,"duration_ms":77746,"reading_group":"no","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["46L55","20M25","20M18"],"pacs":[],"model":"glm-5.2","headline":"Simplicity of Katsura algebras reduced to checking matrix entries","keywords":["Katsura algebras","self-similar groupoids","singular ideals","Steinberg algebras","non-Hausdorff groupoids","simplicity","Kirchberg algebras","algorithmic decidability"],"falsifier":"A pair of matrices (A, B) for which the combinatorial conditions of Theorem 5.7 or 6.14 predict a vanishing singular ideal, but where the C*-algebraic singular ideal is demonstrably nonzero (or vice versa), would break the main equivalence and falsify the simplicity characterization.","tokens_in":28265,"feed_emoji":"🔢","tokens_out":1324,"duration_ms":306282,"temperature":0.7,"pith_summary":"Katsura algebras are C*-algebras built from a pair of integer matrices A and B. They are always nuclear, separable, and satisfy the UCT, but they are simple (and purely infinite) only for certain choices of A and B. The main obstacle to deciding simplicity has been the so-called singular ideal, a structural obstruction that arises specifically when the underlying groupoid is non-Hausdorff. This paper gives a complete characterization of when the singular ideal vanishes, stated entirely in terms of combinatorial conditions on the matrices A and B — specifically, the existence of certain zero paths and the rationality of ratios B_c/A_c along cycles in the graph encoded by A. The characterization covers both the standard Katsura algebra and its faithful quotient, and the authors prove that the vanishing of the algebraic singular ideal (detectable in a purely algebraic Steinberg algebra over any field) is equivalent to the vanishing of the analytic singular ideal in the C*-algebra. They further show that these conditions can be checked by an algorithm running in polynomial space on the input matrices, and provide the first concrete non-Hausdorff examples of non-contracting self-similar groupoids where simplicity is algorithmically decidable.","feed_headline":"Simplicity of Katsura algebras reduced to matrix conditions","feed_subtitle":"Complete characterization of when these C*-algebras are simple, checkable by a polynomial-space algorithm on the input matrices.","key_machinery":"The argument proceeds by viewing the Katsura algebra as the C*-algebra of an ample groupoid arising from a self-similar groupoid action (the KEP-groupoid). The singular ideal is studied via the associated inverse semigroup algebra, using characterizations of tight and singular elements in terms of path-level combinatorics (Propositions 2.4 and 2.5). The reduction from C*-simplicity to algebraic simplicity passes through the gauge-invariant subgroupoid, whose amenability and abelian isotropies allow applying recent results linking algebraic and analytic singular ideals (Theorem 3.3). For the faithful case, the key objects are maximal stable subgroups of finite isotropy groups associated to简单回","core_discovery":"The singular ideal of a Katsura algebra vanishes if and only if certain purely combinatorial conditions on the matrix pair (A, B) hold: for the non-faithful case, every non-faithful isotropy group must either fail to have an infinite path avoiding zero B-edges, or fail to have zero paths extending every finite prefix (Theorem 5.7); for the faithful case, every maximal stable subgroup associated to a simple cycle must have a free orbit on infinite paths (Theorem 6.14). In both cases, the condition is field-independent and equivalent to vanishing of the C*-algebraic singular ideal, yielding a complete simplicity criterion (Corollaries 5.8 and 6.15) checkable in polynomial space.","pith_inferences":["Since Katsura algebras realize all UCT Kirchberg algebras up to K-theory, and the simplicity criterion is now decidable in polynomial space, one could in principle pre-compute a large database of simple vs. non-simple Katsura algebras indexed by small matrix pairs, enabling systematic experimental study of the boundary between simple and non-simple.","The independence of the simplicity criterion from the base field K suggests a deeper rigidity: the combinatorial structure of the groupoid entirely determines the algebraic simplicity across all characteristics, which may point toward a purely order-theoretic or topological explanation that bypasses the field altogether.","The polynomial-space (but not necessarily polynomial-time) complexity bound raises the question of whether the decision problem is PSPACE-complete, which would connect algebraic simplicity to computational complexity classes in a way not previously seen in operator algebras."],"forward_implications":["Given any pair of integer matrices (A, B), one can now algorithmically determine in polynomial space whether the corresponding Katsura algebra is purely infinite simple, making the full UCT Kirchberg algebra classification pipeline constructive for this class.","The equivalence of algebraic and C*-simplicity for these groupoids suggests that the open question of whether algebraic and analytic singular ideals always coincide may be tractable for broad classes of self-similar groupoids beyond the Katsura setting.","The framework of stable subgroups and free-orbit detection may extend to other non-contracting self-similar groupoids, providing tools where the contracting-case machinery of Nekrashevych does not apply.","The explicit non-Hausdorff examples with nontrivial singular ideals (Section 8) provide concrete test cases for further development of non-Hausdorff groupoid C*-algebra theory."],"fun_headline_variants":["Katsura algebra simplicity pinned to matrix pair conditions","When Katsura algebras are simple: a full combinatorial criterion","Simplicity of Katsura algebras decidable in polynomial space","Singular ideal criterion for Katsura algebra simplicity","Katsura algebra simplicity reduces to combinatorial checks on (A, B)"],"cache_read_input_tokens":0,"weakest_assumption_plain":"The bridge between algebraic and analytic simplicity rests on the gauge-invariant subgroupoid being amenable with abelian isotropies, which allows importing recent results about singular ideals. If the amenability transfer from the full groupoid to this subgroupoid has a gap, the equivalence between the checkable algebraic conditions and actual C*-simplicity would not hold.","fun_headline_variants_meta":{"raw":{"variants":["Katsura algebra simplicity pinned to matrix pair conditions","When Katsura algebras are simple: a full combinatorial criterion","Simplicity of Katsura algebras decidable in polynomial space","Singular ideal criterion for Katsura algebra simplicity","Katsura algebra simplicity reduces to combinatorial checks on (A, B)"]},"model":"glm-5.2","effort":"low","cost_usd":0.0,"raw_usage":{"total_tokens":555,"prompt_tokens":478,"completion_tokens":77,"prompt_tokens_details":null},"tokens_in":478,"tokens_out":77,"duration_ms":39518,"temperature":1.0,"reasoning_tokens":null,"cache_read_input_tokens":0,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-07-09T16:44:20.541845+00:00","model_set":{"reader":"glm-5.2"},"falsifier":"A pair of matrices (A, B) for which the combinatorial conditions of Theorem 5.7 or 6.14 predict a vanishing singular ideal, but where the C*-algebraic singular ideal is demonstrably nonzero (or vice versa), would break the main equivalence and falsify the simplicity characterization.","supporting_citations":[],"review_version":1}