{"id":"07655203-11b4-4be2-82c7-efe1688737fb","arxiv_id":"2412.15968","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":8.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"A new proof of the Kirchberg–Phillips theorem that unifies the stably finite and purely infinite classification frameworks and recovers Kirchberg's Geneva theorems as corollaries.","lead":"This mathematics paper gives a new proof of the Kirchberg–Phillips classification theorem for simple nuclear C*-algebras, showing that K-theoretic invariants completely determine the algebras. The proof uses one unified framework for both the finite-trace and purely infinite cases, and it derives two deep structure theorems, formerly used as inputs, as consequences.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The central independence claim depends on Theorem 4.19, imported from [6, Thm 5.15] with only a sketch of how Z-stability is replaced; if that replacement is not independently verifiable, the new-proof claim lacks support.","rationale":"Reader's weakest assumption is the same as mine. The paper has genuine new content: the state-kernel extension, pure largeness of e_B, the de-unitization trick, and the intertwining argument are all presented in detail. The concern is not that these are wrong; it is that the paper's most important advertised novelty—independence from Geneva theorems—cannot be certified from the text alone. In particular, Theorem 4.19 is the place where the purely infinite argument most differs from [6], and the difference is compressed into one sentence. Since [6] is by the same author group and may itself assume the stably finite classification framework only, a hidden use of Z-stability or of the O∞-absorption theorem would be a genuine circularity. I therefore retain CONDITIONAL. I would not change the reader's verdict: the proof is plausible and well organized, but the cited dependency must be community-verified before the independence claim can be accepted.","tokens_in":42768,"tokens_out":20004,"duration_ms":181371,"concrete_test":"Perform an independent audit of [6, Theorem 5.15]. List every line of its proof that uses Z-stability or Z-stability of I; for each such line, write the corresponding argument when I is only assumed separable, simple, purely infinite and stable, using Lemma 4.18 and Loreaux–Ng [26, Theorem 2.5]. If some line uses Z ≅ Z⊗Z to construct central sequences or asymptotic unitaries and no replacement is supplied, Theorem 4.19 is not established. A clean test: re-derive Theorem 4.19 from the tools stated in §4 without consulting [6, §5.4]; if the derivation succeeds, the concern is resolved; if it stalls at the 'second half', the paper must either prove the theorem or weaken the independence claim.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The advertised result is not just Kirchberg–Phillips, but a proof of it that avoids Kirchberg's Geneva Theorems. The uniqueness engine is Theorem 4.19, whose proof is not in the paper: it says 'the exact same proof as that of [6, Theorem 5.15] works here' after replacing Z-stability by K1-injectivity from Loreaux–Ng. In [6], Z-stability is a standing hypothesis and is used in the second half of the proof; the purely infinite hypotheses of Theorem 4.19 do not by themselves supply Z-stability unless one invokes O∞-absorption, which is one of the very results the paper claims to recover as a corollary. The one-sentence removal of the 'tensorial factor Z' is therefore load-bearing and unverified. If the step cannot be reconstructed without hidden use of O∞-absorption or the Kirchberg–Phillips theorem, then Theorems 6.3, 6.4 and 6.16 lose their foundation and the central independence claim fails, even though the theorem statement may be true. This is an external-provenance concern, not an internal contradiction; I found no flaw in the new state-kernel construction itself.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper adapts the Carrión–Gabe–Schafhauser–Tikuisis–White classification framework, originally developed for stably finite Z-stable algebras, to the purely infinite setting. It constructs a reduced state-kernel extension using a strongly faithful state on a simple purely infinite C*-algebra, proves that the extension is purely large, and then develops KK-existence and KL-uniqueness results, a separabilization procedure, and a classification of lifts. These are combined to prove a classification of unital embeddings into purely infinite C*-algebras and hence a new proof of the Kirchberg–Phillips theorem under the UCT. The paper also derives the nuclear versions of Kirchberg's Geneva theorems (O2-embedding, O2-absorption, O∞-absorption) as corollaries, without using them as hypotheses.","tokens_in":42945,"tokens_out":19276,"duration_ms":182596,"significance":"If the technical results imported from [6] and [16] are valid and independent of the classification results being reproved, this is a substantial contribution: it unifies the stably finite and purely infinite classification frameworks and provides a genuinely new route to the Kirchberg–Phillips theorem in which Kirchberg's Geneva theorems are consequences rather than ingredients. The state-kernel construction, the proof that the reduced state-kernel extension is purely large, and the de-unitization argument are original and appear to be carried out in considerable detail. The paper is also honest about the provenance of several key tools, explicitly quoting the relevant theorems from [6] and [16]. However, the central independence claim cannot currently be verified from the manuscript because the main uniqueness theorem is not proved in the paper; this is a load-bearing external-provenance concern rather than an internal inconsistency.","major_comments":[{"comment":"The proof of the central uniqueness theorem is not contained in the paper: the text says that by using Lemma 4.18, 'the exact same proof as that of [6, Theorem 5.15] works here' and that the only difference is removing the tensorial factor Z. This is load-bearing: Theorem 4.19 is the uniqueness engine for Theorem 6.3 and hence for the classification theorems in Section 6, and the paper's advertised claim that it avoids Kirchberg's Geneva Theorems depends on this replacement not secretly using O∞-absorption or the Kirchberg–Phillips theorem. The one-sentence removal of Z-stability is not sufficient for the reader to verify that the second half of the proof in [6] remains valid without Z-stability and without the very theorems the paper seeks to recover. I recommend that the authors provide a complete proof of Theorem 4.19, or a detailed account of how the proof of [6, Theorem 5.15] is modified, explicitly checking that no step invokes the purely infinite classification or O∞-absorption.","section":"§4.3, Theorem 4.19"},{"comment":"The independence claim also relies on several quoted results from the authors' own preprint [6] and from [16]. Theorem 4.17 is quoted from [6, Theorem 5.14] and is used in the existence part of Theorem 6.3, while Lemma 6.11 is quoted from [16, Proposition 12.24] and is used in the uniqueness proof of Theorem 6.16. Since [6] is a preprint and [16] is a classification memoir for O∞-stable algebras, the paper should state for each such quoted result whether it is available independently of Kirchberg's Geneva Theorems and the Kirchberg–Phillips theorem, ideally by giving a proof or a precise reference to a published source with that independence established. Without this, the claim that the proof 'does not rely on Kirchberg's Geneva Theorems' is not checkable.","section":"§4.3 and §6.2, Theorems 4.17 and 6.16"}],"minor_comments":[{"comment":"In the displayed definition of the Cuntz sum, the second term is written as s1 φ(a) s1* again; it should be s2 ψ(a) s2*.","section":"§4.1, Definition 4.1"},{"comment":"The phrase 'unital, separable, unital, simple' contains a duplicated word 'unital'; the first occurrence should be removed.","section":"§6.2, Corollary 6.20"},{"comment":"There is a spelling typo: 'faithul' should be 'faithful'.","section":"Remark 3.9"},{"comment":"There are several spacing and OCR-style artifacts (for example 'W e', 'fi nite', 's imply') in the text; a careful proofreading pass would improve readability.","section":"Abstract and throughout"},{"comment":"The final step 'since we are working in Bω, this means that they are unitarily equivalent' should cite the standard ε-test or a reference, because approximate unitary equivalence in Bω does not imply unitary equivalence without an argument using separability of the domain.","section":"§6.3, Theorem 6.13"}],"recommendation":"major_revision","confidential_remarks":"For the editor: the manuscript is mathematically substantial and the novel parts are generally well argued, but the central independence claim rests on Theorem 4.19, whose proof is only asserted to be 'the exact same proof' as a result in the authors' own preprint [6]. This is a verifiability problem that should be resolved before publication. I would not reject the paper on this basis, because the gap may be fixable by including a full proof or by clarifying that [6] is accepted and independent. I would also advise the editor that the paper relies heavily on [6] and [16]; if the journal's policy discourages heavy reliance on unpublished preprints, that should be addressed."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Seb, this is a serious paper and worth a careful read. The core new idea is the reduced state-kernel extension: replace traces with a single strongly faithful state, restrict the sequence algebra to the C*-subalgebra where the state-kernel is an ideal, and end up with B(H) as the quotient in the purely infinite case. That lets them run the CGSTW machinery without any trace data and without type-III von Neumann classification. The proof is detailed: pure largeness of the extension, separabilization, classification of lifts, and the intertwining. Recovering the Geneva theorems as corollaries is genuinely new, and the paper is honest that it only handles nuclear C*-algebras.\n\nThe soft spot is exactly where the stress-test lands. Theorem 4.19, the KK-/KL-uniqueness engine, is not proved; it says 'the exact same proof as [6, Theorem 5.15]' after swapping Z-stability for K1-injectivity by Loreaux–Ng. That is load-bearing. The paper asserts Z-stability is automatic in this setting, but the assertion is not justified in the text, and without seeing the second half of the [6] proof one cannot tell whether replacing the tensorial factor Z introduces a clandestine use of O∞-absorption. The KK-existence theorem is also imported from [6]. These are self-citations, which is not itself a flaw, but the independence claim of the paper rests on those quoted results being independent of Kirchberg–Phillips. That is currently unverified by the paper.\n\nThere is also a statement issue: Theorem 6.17 states B as 'unital, simple, separable and purely infinite' with no nuclearity. If that is not a typo, it's a much stronger claim than Kirchberg–Phillips and likely false for non-nuclear B. Worth checking before the paper circulates further.\n\nOverall, the architecture is sound and the new parts are proved in detail. The uncertainty is concentrated in one imported proof. This deserves a serious referee, but the referee should demand a full proof of Theorem 4.19 or a precise citation to a version of [6] with the purely infinite adaptation.","headline":"A serious unified proof of Kirchberg–Phillips with a genuinely new state-kernel construction; the independence claim hinges on one imported uniqueness theorem that is not proved here.","tokens_in":43526,"tokens_out":9080,"would_cite":true,"duration_ms":80135,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["46L35","46L80"],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper proves the Kirchberg–Phillips classification by a unified K-theoretic framework that works for both finite and purely infinite C*-algebras.","keywords":["C*-algebra classification","Kirchberg–Phillips theorem","purely infinite C*-algebras","K-theory","KK-theory","universal coefficient theorem","state-kernel extension","Cuntz algebras"],"falsifier":"Check the proofs of the quoted KK-existence and uniqueness theorems for any use of the O2-embedding theorem, O∞-absorption, or the Kirchberg–Phillips theorem itself; alternatively, find two unital UCT Kirchberg algebras with the same total K-theory and same unit class that are not isomorphic, which would disprove Theorem 6.18.","tokens_in":42500,"feed_emoji":"∞","tokens_out":7008,"duration_ms":57311,"temperature":0.7,"pith_summary":"The paper claims that Kirchberg–Phillips, the classification of unital simple purely infinite nuclear C*-algebras by K-theory, can be proved with the same abstract machinery that was recently developed for stably finite algebras. The proof replaces traces, which do not exist in the purely infinite case, with a single strongly faithful state, and uses the resulting reduced state-kernel extension to run the same KK-theoretic existence and uniqueness arguments. If the proof is right, Kirchberg's Geneva theorems become consequences of classification rather than prerequisites.","feed_headline":"New proof of Kirchberg–Phillips needs no Geneva theorems","feed_subtitle":"A state-kernel extension classifies purely infinite C*-algebras by K-theory and turns O2 and O∞ absorption into corollaries.","key_machinery":"The central object is the reduced state-kernel extension. Given a strongly faithful state ρ on B, the paper forms 0 → J_{B,ρ} → S_{B,ρ} → π_ρ(B)'' → 0, where J_{B,ρ} is the ideal of sequences whose ρ-norm tends to zero and S_{B,ρ} is the C*-subalgebra of the sequence algebra B∞ consisting of ρ-Cauchy sequences. This replaces the trace-kernel extension used in the stably finite setting. Choosing ρ so that the quotient is B(H) reduces the von Neumann side to Voiculescu's theorem, and the extension is shown to be purely large, so the Elliott–Kucerovsky absorption theorem makes its Busby map absorbing. The classification of lifts then carries the argument.","core_discovery":"The central statement is Theorem 6.17: for a unital, separable, nuclear C*-algebra A satisfying the UCT and a unital, simple, separable, purely infinite C*-algebra B, every map of total K-theory sending the class of the unit of A to the class of the unit of B is realized by a unital injective ∗-homomorphism from A to B, and any two such maps are approximately unitarily equivalent. In the special case of Kirchberg algebras, this gives A ≅ B if and only if (K_*(A), [1_A]_0) ≅ (K_*(B), [1_B]_0). The proof claims independence from Kirchberg's Geneva theorems and recovers them as corollaries.","pith_inferences":["The same state-kernel construction may extend to non-simple O∞-stable algebras, since pure largeness rather than traces encodes the ideal structure.","A natural test is whether reindexing arguments can be restored to upgrade the uniqueness statement from approximate unitary equivalence in Bω to unitary equivalence in B∞.","Because the paper relies on unproduced results from the companion preprint for its KK-existence and uniqueness theorems, the fastest check of the independence claim is to verify those proofs for hidden uses of Kirchberg–Phillips or the Geneva theorems.","The KK-rigidity corollary suggests a path toward purely infinite classification results without the UCT, in the style of the stably finite case."],"forward_implications":["Two unital Kirchberg algebras satisfying the UCT are isomorphic exactly when their total K-theory together with the K0-class of the unit agree (Theorem 6.18).","Every unital, separable, nuclear C*-algebra embeds unitally into O2 (Corollary 6.19).","For unital simple nuclear A, A ⊗ O2 ≅ O2 and, for purely infinite A, A ⊗ O∞ ≅ O∞, with O2 and O∞ strongly self-absorbing (Corollaries 6.20 and 6.23).","A unital ∗-homomorphism between Kirchberg algebras that is a KK-equivalence is approximately unitarily equivalent to an isomorphism (Proposition 6.21).","The purely infinite classification no longer needs the Z-stability machinery or the corona factorization property used in the stably finite proof."],"supporting_citations":[{"why":"Supplies the abstract framework, the KK-existence theorem, the KK/KL-uniqueness theorem, and the separabilization lemmas that the paper quotes wholesale.","marker":"[6]"},{"why":"Gives the absorption theorem equating pure largeness with absorbing Busby maps, used to turn the reduced state-kernel extension into an absorbing map.","marker":"[14]"},{"why":"Provides standard facts on purely infinite algebras, ultrapower simplicity, and the Cuntz subequivalence techniques used throughout.","marker":"[30]"},{"why":"Supplies the Choi–Effros lifting theorem and nuclearity facts used repeatedly in the proofs.","marker":"[5]"},{"why":"Gives the stability criterion used to show the ideal in the separabilized extension is stable and to relate separable stability with stability.","marker":"[20]"},{"why":"Provides the K1-injectivity result needed in the KK/KL-uniqueness theorem without Z-stabilization.","marker":"[26]"},{"why":"Supplies the observation that inclusion of a full hereditary subalgebra induces a KK-isomorphism, used in Lemma 6.11.","marker":"[16]"},{"why":"Motivates the non-UCT corollary by showing how KK-equivalences can force approximate unitary equivalence to isomorphisms.","marker":"[33]"}],"fun_headline_variants":["Unified proof classifies nuclear C*-algebras without Geneva theorems","New route to Kirchberg–Phillips: no Geneva needed","State-kernel method unifies C*-algebra classification","K-theory settles Kirchberg–Phillips, Geneva as corollary","Geneva theorems become corollaries in unified C*-algebra classification"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The proof rests on the assertion, taken from the companion preprint, that its KK-existence and KK/KL-uniqueness theorems and separabilization lemmas are proved without using Kirchberg's Geneva theorems or the Kirchberg–Phillips theorem; if that assertion is false, the paper's claim of a new independent proof collapses.","fun_headline_variants_meta":{"raw":{"variants":["Unified proof classifies nuclear C*-algebras without Geneva theorems","New route to Kirchberg–Phillips: no Geneva needed","State-kernel method unifies C*-algebra classification","K-theory settles Kirchberg–Phillips, Geneva as corollary","Geneva theorems become corollaries in unified C*-algebra classification"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001224,"raw_usage":{"total_tokens":4966,"prompt_tokens":814,"completion_tokens":4152,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":430,"completion_tokens_details":{"reasoning_tokens":4064}},"tokens_in":430,"tokens_out":4152,"duration_ms":25409,"temperature":1.0,"reasoning_tokens":4064,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T10:56:07.777167+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Check the proofs of the quoted KK-existence and uniqueness theorems for any use of the O2-embedding theorem, O∞-absorption, or the Kirchberg–Phillips theorem itself; alternatively, find two unital UCT Kirchberg algebras with the same total K-theory and same unit class that are not isomorphic, which would disprove Theorem 6.18.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Gives the absorption theorem equating pure largeness with absorbing Busby maps, used to turn the reduced state-kernel extension into an absorbing map."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the Choi–Effros lifting theorem and nuclearity facts used repeatedly in the proofs."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Gives the stability criterion used to show the ideal in the separabilized extension is stable and to relate separable stability with stability."},{"cited_title":"Loreaux and P","cited_arxiv_id":null,"evidence_quote":"Provides the K1-injectivity result needed in the KK/KL-uniqueness theorem without Z-stabilization."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the observation that inclusion of a full hereditary subalgebra induces a KK-isomorphism, used in Lemma 6.11."},{"cited_title":"KK-rigidity of simple nuclear C*-algebras","cited_arxiv_id":"2408.02745","evidence_quote":"Motivates the non-UCT corollary by showing how KK-equivalences can force approximate unitary equivalence to isomorphisms."}],"review_version":1}