{"id":"21888700-746f-463e-ae9f-7b2d4eb59d5f","arxiv_id":"2510.19735","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Simplicity of Steinberg algebras of contracting self-similar groupoids is characterized by recurrent subgroups of the nucleus, and coincides with simplicity of the reduced C*-algebra.","lead":"This paper gives a criterion for when certain algebras built from self-similar groupoids have no nontrivial ideals, and shows the criterion is the same for the algebraic and operator-algebraic versions in the contracting case. The test runs on a finite set called the nucleus, so it can be applied mechanically to examples like the Basilica group.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"C*-simplicity bridge (Cor. 9.3) rests on unproved equality G(S,Γ^ω)=G_T(S); [34] covers only self-similar groups, and the identification fails for graphs with sinks.","rationale":"The reader's weakest assumption identifies the same load-bearing gap: the C*-algebra simplicity result depends on identifying G(S,Γ^ω) with the tight groupoid G_T(S), and the paper gives only a reference for self-similar groups, not groupoids. This is a genuine concern because Corollary 9.3 is one of the paper's headline claims and the identification is not a formal consequence of the definitions as written. The unit-space issue for graphs with sinks makes the gap concrete rather than merely expository. The algebraic core—Corollary 8.7 and its supporting Sections 5–8—does not rest on this identification, so the algebraic simplicity criterion is not jeopardized. The verdict should remain conditional: the paper is acceptable if the missing identification is supplied, with appropriate hypotheses ruling out or handling sinks.","tokens_in":26277,"tokens_out":27328,"duration_ms":239861,"concrete_test":"Take the trivial self-similar groupoid on the graph with one edge u→v and no edge out of v. Compute the tight groupoid G_T(S) from the tight filters on E(S)={u,v,0} and compare its unit space with Γ^ω=∅. If G_T(S) is nonempty, the unqualified identification asserted in Section 9 is false as stated. If G_T(S) is also empty, repeat the comparison for a self-similar groupoid whose graph has a sink and nontrivial isotropy; in either case, the paper needs an explicit argument, not a citation to the self-similar group case.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Section 9's Corollaries 9.2 and 9.3 apply [4] to the groupoid G=G(S,Γ^ω), said to be the tight groupoid of S. The only justification is the sentence \"It turns out that the action of S on Γ^ω can be identified with the standard action of S on the space of tight characters of S... see [34] for this identification in the self-similar group case.\" That reference treats self-similar groups on a one-sided shift, not self-similar groupoids on arbitrary row-finite graphs. The identification is not automatic: G_T(S) has unit space the tight spectrum of E(S)={pp*}∪{0}, while G(S,Γ^ω) uses only infinite paths. For a graph with a sink the two unit spaces need not agree; e.g. the trivial groupoid on a single edge u→v with no edge out of v has Γ^ω=∅, but the idempotents u,v yield candidate tight characters. Section 9 imposes no no-sink hypothesis. Even when every vertex emits an edge, one still needs a proof that every tight character of E(S) is a boundary point and that the germ relation for the S-action on Γ^ω coincides with the tight-filter action. Without this, Corollary 9.3 cannot be derived from [4, Corollaries 4.6, 4.8]. The algebraic simplicity criterion (Cor. 8.7) is independent and appears supported.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper develops a simplicity criterion for Steinberg algebras of self-similar groupoids. It constructs the associated inverse semigroup S(G,Γ*), characterizes when this inverse semigroup is congruence-free (Theorem 3.1), and identifies the tight ideal with the Cuntz-Krieger ideal (Proposition 4.3). Using the Steinberg–Szakács result that KS/I_tight is simple iff I_tight = I_ess (Theorem 4.4), it reduces the question to a comparison of ideals. Section 5 collapses a self-similar groupoid to its skeleton, Section 6 shows a difference of ideals is supported on the groupoid algebra KG, and Section 7, in the contracting case, further localizes the difference to spans of recurrent subgroups KH_p. Section 8 provides the final combinatorial criterion in terms of the graph Δ and synchronized vertices (Theorem 8.6 and Corollary 8.7). Section 9 claims that for a contracting self-similar groupoid the complex Steinberg algebra is simple iff the reduced C*-algebra is simple (Corollary 9.3), by identifying the groupoid of germs of the action on Γ^ω with the tight groupoid of S. Section 10 applies the results to a Basilica-type groupoid and to a groupoid containing Grigorchuk and Grigorchuk–Erschler subgroups.","tokens_in":26662,"tokens_out":9030,"duration_ms":82130,"significance":"If the algebraic results stand—and they appear to be proved in detail—the paper gives a concrete, checkable criterion for simplicity of Steinberg algebras of self-similar groupoids, extending the self-similar group results of Steinberg–Szakács and Gardella–Nekrashevych–Steinberg–Vdovina. The collapse-to-skeleton reduction (Section 5) and the Δ-graph criterion (Section 8) are useful tools, and the worked examples show the method in action. The claimed C*-algebra equivalence would be a natural and valuable companion result. However, the C* bridge in Section 9 rests on an unproved identification that is known to fail in the presence of sinks; this is a load-bearing gap in the main advertised theorem. The algebraic core should still be considered a substantial contribution once the C* section is repaired or restricted appropriately.","major_comments":[{"comment":"The paper asserts without proof that G(S,Γ^ω), the groupoid of germs of the action of S on infinite paths, coincides with the tight groupoid G_T(S). The only reference given, [34], proves this identification in the self-similar group case, not for self-similar groupoids acting on arbitrary row-finite graphs. The identification is not automatic: the unit space of G_T(S) is the tight spectrum of E(S)={pp*}∪{0}, which for a graph with a sink includes characters associated to finite paths ending at that sink, whereas Γ^ω contains only infinite paths. No no-sink hypothesis is imposed in Section 9. For example, the trivial self-similar groupoid on the single edge u→v with v a sink has Γ^ω=∅, but the idempotents v and ee* admit tight characters, so G(S,Γ^ω) cannot equal G_T(S). Consequently Corollaries 9.2 and 9.3 do not follow from [4, Corollaries 4.6, 4.8] as written. The authors should eithe","section":"Section 9, paragraph beginning 'It turns out that the action of S on Γ^ω...' and Corollaries 9.2, 9.3"},{"comment":"The main advertised theorem—that simplicity of the reduced C*-algebra of a contracting self-similar groupoid coincides with simplicity of the Steinberg algebra—is stated without the additional hypotheses (e.g. no sinks, or type (CF)) under which the Section 9 argument might work. Even if the identification G(S,Γ^ω)=G_T(S) can be established for a suitable boundary path space, the statement needs to be reformulated with the correct hypotheses and the proof supplied. As it stands, the C*-algebra claim is not established for the generality claimed.","section":"Abstract and Corollary 9.3"}],"minor_comments":[{"comment":"The statement is clearly a typo: both sides read 'KS(L,Λ*)'. The first side should be 'KS(G,Γ*)' and the second 'KS(L,Λ*)'.","section":"Corollary 5.5"},{"comment":"The notation 'I_ess ⊊ I_tight' is backwards; since I_tight ⊆ I_ess, the intended condition is 'I_tight ⊊ I_ess' (or equivalently I_ess \\ I_tight nonempty). This appears in the theorem statement and in the first sentence of its proof.","section":"Theorem 8.6 and its proof"},{"comment":"The symbol p is used both for an infinite path and for finite prefixes of it (e.g. '{[p n p*,p] : n∈N, p is a prefix of p}'). This makes the proof harder to follow; renaming the infinite path, say ξ, would clarify.","section":"Section 9, Proposition 9.1"},{"comment":"Typo: 'bouqet' should be 'bouquet'.","section":"Example 10.1"},{"comment":"The paper uses the same symbol * for path adjoints, inverse-semigroup adjoints, and C*-algebra adjoints. The authors acknowledge this at the start, but a short glossary or a consistent font distinction would improve readability.","section":"Throughout"}],"recommendation":"major_revision","confidential_remarks":"The algebraic core of the paper (Sections 3–8) is solid and appears to be a genuine contribution. The only serious obstruction is the unproved and, in the presence of sinks, false identification in Section 9. If the authors can either prove the tight-groupoid identification for the full boundary path space of a row-finite graph or restrict the C*-statement to the no-sink case with a complete proof, the paper would be suitable for publication. I would not recommend rejection based on the algebraic results, but the current statement of Corollary 9.3 cannot be accepted without repair."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Read this for Sections 3–8. Aakre gives a complete criterion for simplicity of the Steinberg algebra of a self-similar groupoid, genuinely extending the self-similar group results of Steinberg–Szakács and Gardella–Nekrashevych–Steinberg–Vdovina to groupoids acting on row-finite graphs. The collapse-to-skeleton construction is a real new tool, and the Δ-graph algorithm turns the criterion into something checkable. The examples—especially the Basilica-type and Grigorchuk/Grigorchuk–Erschler-type groupoids—actually exercise the machinery rather than just decorating it. The algebra part is proved in detail and looks sound.\n\nThe soft spot is Section 9. Corollary 9.3 claims that simplicity of the complex Steinberg algebra coincides with simplicity of the reduced C*-algebra for contracting self-similar groupoids. The derivation applies [4] to the groupoid of germs G(S,Γ^ω), which the paper says “turns out” to be the tight groupoid G_T(S). The cited reference [34] proves this identification only for self-similar groups, not groupoids. That matters: the identification is not automatic, since the unit space of G_T(S) is the tight spectrum of the idempotent semilattice while G(S,Γ^ω) uses only infinite paths. For graphs with sinks these unit spaces can disagree, and Section 9 imposes no no-sink hypothesis. So as written, Corollaries 9.2 and 9.3 do not follow from [4]. This is a genuine gap in the C*-part, though it does not affect the algebraic core.\n\nMinor: Corollary 5.5 has a typo—it says the essential and tight ideals of KS(L,Λ*) coincide with themselves; one side should be KS(G,Γ*). Easy fix. The note that Theorem 3.1 appeared in the author's master's dissertation is harmless because the theorem is proved in full here.\n\nThis paper is for people working on Leavitt path algebras, Nekrashevych algebras, and étale groupoid algebras; they will find the algebraic criterion and the collapse construction useful regardless of what happens to the C*-bridge. The C*-claim, once fixed or properly conditioned, would be the headline—but right now it is conditional.\n\nYes, send it to peer review. The algebraic results are significant enough to merit serious refereeing. A referee should press for a proof of the identification in Section 9 (or a corrected hypothesis), but the paper deserves that effort.","headline":"The algebraic simplicity criterion for self-similar groupoids (Sections 3–8) is a real, checkable contribution; the C*-algebra bridge in Section 9 rests on an unproved identification and needs work before the headline claim can be trusted.","tokens_in":27101,"tokens_out":2832,"would_cite":true,"duration_ms":24746,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["16S38","20M18","46L05"],"pacs":[],"model":"deepseek-v4-flash","headline":"For a self-similar groupoid of type (CF), the Steinberg algebra is simple exactly when certain projections on recurrent subgroups of the nucleus have trivial common kernel; in the contracting case this condition also controls simplicity of","keywords":["self-similar groupoids","Steinberg algebras","inverse semigroup algebras","tight ideal","essential ideal","reduced C*-algebras","nucleus of a self-similar groupoid","simplicity criterion"],"falsifier":"Exhibit a self-similar groupoid (not a group) for which the groupoid of germs G(S,Γ^ω) is not isomorphic to the tight groupoid G_T(S). This can be tested by finding a germ [s,p] whose source and range are not tight characters of S; if such a germ exists, the identification in Section 9 breaks and Corollary 9.3 would not be justified.","tokens_in":26164,"feed_emoji":"🧩","tokens_out":9447,"duration_ms":69299,"temperature":0.7,"pith_summary":"The paper aims to characterize exactly when the Steinberg algebra of a self-similar groupoid — a discrete algebra generalizing Leavitt path algebras and Nekrashevych algebras — is simple, and to show that in contracting cases this matches simplicity of the reduced C*-algebra. The main result reduces the question to a finite algebraic test: build a small graph from the nucleus, find its synchronized vertices, and check whether certain projections associated to recurrent subgroups have trivial common kernel. The paper also proves that one may collapse a self-similar groupoid to its skeleton acting on a strongly connected graph without changing the simplicity of the algebra. A key bridge between the discrete and analytic worlds is the identification of the groupoid of germs of the boundary action with the tight groupoid, which is cited from prior work for groups but not fully justified for groupoids.","feed_headline":"Simplicity of self-similar groupoid algebras reduces to a finite check","feed_subtitle":"Using the inverse semigroup, simplicity is decided by kernels of projections on recurrent subgroups.","key_machinery":"The load-bearing object is the inverse semigroup S(G,Γ*) built from the self-similar groupoid's action on paths, and the contracted algebra K0S; simplicity is governed by the equality of the tight ideal I_tight (generated by Cuntz-Krieger relations) with the essential ideal I_ess (singular elements). The paper's algorithm uses the nucleus N of a contracting groupoid, the recurrent subgroups H_p = {g : g(p)=p, g|_p = g} ⊆ N, a graph ∆ whose vertices are subsets I_q of the nucleus's cyclic part, and coset projections π_{p,q}: KH_p → K[H_p/(H_p∩I_q)]. The equivalence π_{p,q}(a)=0 ⇔ aq=0 in the algebra lets one decide membership in I_ess by checking only synchronized vertices of ∆.","core_discovery":"The central claim is that for a self-similar groupoid of type (CF) — meaning every vertex emits at least two edges and the action mixes vertices suitably — the contracted inverse semigroup algebra modulo its tight ideal (equivalently, the Steinberg algebra of the tight groupoid) is simple if and only if, for every recurrent subgroup H_p contained in the nucleus, the intersection of the kernels of the projections π_{p,q} over all synchronized vertices I_q with s(q)=s(p) is zero. In the contracting case, the essential ideal and tight ideal can differ only via an element of KH_p for some recurrent subgroup, leading to this criterion. Moreover, under the same contracting hypothesis, simplicity o","pith_inferences":["The identification of the groupoid of germs with the tight groupoid is only cited for self-similar groups; a proof or counterexample for groupoids would settle whether the C*-algebra result is unconditional.","The skeleton collapse suggests that the subtleties unique to groupoids — multiple orbits, non-trivial isotropy between vertices — can often be bypassed, so future work might transfer known results for self-similar groups to groupoids with minimal effort.","The characteristic sensitivity seen in the multispinal-style example hints that simplicity of the Steinberg algebra over a field K encodes modular information about the recurrent subgroups; one might predict a connection to the K-theory or homology of the groupoid.","Because ∆ is finite in the contracting case, the kernel-intersection test is algorithmic; implementing it for a library of examples could reveal new contracting groupoids with prescribed simplicity properties."],"forward_implications":["The criterion gives a concrete algorithm: compute the nucleus, the recurrent subgroups, the graph ∆, and the synchronized vertices, then check the kernel intersections; all data is finite for contracting groupoids.","Because the family of Steinberg algebras includes Leavitt path algebras and Nekrashevych algebras, the result unifies and extends simplicity criteria for those well-studied classes.","In the contracting case, the same discrete condition decides whether the reduced C*-algebra is simple, so analytic simplicity can be established by purely algebraic computations.","Collapsing to the skeleton shows that simplicity is invariant under the collapse to a self-similar group acting on a strongly connected graph, so one may simplify computations without changing the answer.","The examples illustrate that the method can distinguish cases where simplicity holds over every field (a Basilica-based groupoid) from cases where simplicity fails in every characteristic (a multispinal-style weaving of two self-similar groups)."],"fun_headline_variants":["Kernels of projections decide simplicity in self-similar groupoids","Finite kernel check settles simplicity of groupoid algebras","Self-similar groupoid simplicity: only recurrent kernels matter","For contracting groupoids, C*-algebra simplicity is a finite check","Simplicity of self-similar groupoid algebras? Check kernel intersections"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The bridge from the discrete Steinberg algebra to the C*-algebra relies on the identification of the groupoid of germs of the inverse semigroup's action on the boundary with the tight groupoid; the paper cites this for self-similar groups but supplies no proof for self-similar groupoids, so if that identification fails in the groupoid case the C*-algebra conclusion would not follow.","fun_headline_variants_meta":{"raw":{"variants":["Kernels of projections decide simplicity in self-similar groupoids","Finite kernel check settles simplicity of groupoid algebras","Self-similar groupoid simplicity: only recurrent kernels matter","For contracting groupoids, C*-algebra simplicity is a finite check","Simplicity of self-similar groupoid algebras? Check kernel intersections"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001185,"raw_usage":{"total_tokens":4685,"prompt_tokens":657,"completion_tokens":4028,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":401,"completion_tokens_details":{"reasoning_tokens":3942}},"tokens_in":401,"tokens_out":4028,"duration_ms":24600,"temperature":1.0,"reasoning_tokens":3942,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-04T08:35:13.722749+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Exhibit a self-similar groupoid (not a group) for which the groupoid of germs G(S,Γ^ω) is not isomorphic to the tight groupoid G_T(S). This can be tested by finding a germ [s,p] whose source and range are not tight characters of S; if such a germ exists, the identification in Section 9 breaks and Corollary 9.3 would not be justified.","supporting_citations":[],"review_version":1}