{"id":"41d01564-d95b-460d-8703-1b7a0d6a1e1d","arxiv_id":"1908.05668","paper_version":3,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"The paper constructs a symmetric monoidal closed model category of Gamma-spaces whose fibrant objects are coherently commutative monoidal quasi-categories, and proves Quillen equivalences with normalized Gamma-spaces and with Lurie's symmetric monoidal quasi-categories.","lead":"This paper constructs a new model category for symmetric monoidal higher categories, built from Gamma-spaces, and proves it is equivalent to the standard coCartesian fibration model. If correct, it gives higher-algebra workers a closed monoidal homotopy theory for coherently commutative monoidal quasi-categories.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Theorem 5.18's pushout-product proof applies Proposition 5.13 to non-cofibrant sources; the central monoidal-closure claim is not established as written.","rationale":"The reader's weakest assumption identifies the unproved statement in Theorem 4.14 that f⊗Γ^n preserves (acyclic) cofibrations. I agree this is a gap in the text, but it is easily filled: the right adjoint is precomposition with n+∧-, which manifestly preserves degreewise Joyal fibrations and acyclic fibrations, so -⊗Γ^n is left Quillen. The more serious written gap is in Theorem 5.18, where Proposition 5.13 is applied without the cofibrancy hypothesis. A standard cellular reduction would repair it, but the paper does not supply it. The false ancillary assertion that fibrations between fibrant objects are strict JQ-equivalences (Theorems 5.11 and 7.16) is not used in the main arguments; the direction actually needed, that strict fibrations between fibrant objects are fibrations in the localized model, is standard. The final comparison with symmetric monoidal quasi-categories relies on the cited relative-nerve equivalence and [Sha, Lemma 3.12]; I did not find an independent gap there beyond the usual dependence on cited preprints. Overall the central construction appears plausible and likely correct, but the written proof of the monoidal closure theorem is incomplete. That supports the reader's CONDITIONAL verdict without moving it.","tokens_in":41847,"tokens_out":38528,"duration_ms":401695,"concrete_test":"Rewrite the acyclicity half of Theorem 5.18 using the cofibrantly generated small-object argument: let I and J be the generating cofibrations and generating acyclic cofibrations of the JQ model structure, and verify directly that i□j is an acyclic JQ-cofibration for every i∈I and j∈J, using Theorem 4.10 and the adjunction -⊗Γ^n ⊣ Map(Γ^n,-). Then close under pushout, transfinite composition, and retract in each variable to obtain the theorem for all i,j. A minimal pair to check is i : Γ^n×∅ -> Γ^n×Δ^0 and j an inner-horn inclusion Γ^m×Λ^r_s -> Γ^m×Δ^r. If the generating-pair verification succeeds, the concern is repaired; if any generating pair fails, Theorem 5.18 is false.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central monoidal closure claim (Theorem 5.18) is not proved as written. To show that i□j is a JQ-equivalence when j is an acyclic JQ-cofibration, the proof invokes Proposition 5.13. That proposition requires the cofibration to be between cofibrant objects. For i a generating JQ-cofibration and j an arbitrary acyclic JQ-cofibration, the source of i□j is U⊗Z ∐_{U⊗Y} V⊗Y; Y and Z are not assumed cofibrant, and the monoidal structure itself is exactly what is being proved, so this pushout need not be cofibrant. Proposition 5.13 therefore does not apply. The subsequent class S is closed under pushout, transfinite composition, and retract in the first variable only, and the 'same argument' for the second variable repeats the same gap. This is not the unproved 'one can show' preservation statement in Theorem 4.14: that statement is true and follows immediately from the adjunction -⊗Γ^n ⊣ Map(Γ^n,-), since precomposition by n+∧- preserves degreewise Joyal fibrations and acyclic fibrations. The gap in Theorem 5.18 needs a cellular reduction to generating acyclic cofibrations, which the text does not provide. Since Theorem 5.18 is the paper's main monoidal-closure result, the central claim is conditional on a missing standard argument.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper constructs a strict JQ-model structure on the category of Γ-spaces, localizes it at the Segal maps to obtain a model category of coherently commutative monoidal quasi-categories, and proves that this localized model category is symmetric monoidal closed under Day convolution. It then compares this model with a normalized version on pointed Γ-spaces, with a marked version on Γ-spaces in marked simplicial sets, and with Lurie's symmetric monoidal quasi-categories via the relative nerve and its left adjoint. The central claims are the existence of the JQ-model category, its monoidal closure, and a Quillen equivalence with the localized model category on marked coCartesian fibrations over N(Γ^op).","tokens_in":42107,"tokens_out":7741,"duration_ms":79972,"significance":"If the advertised results hold, the paper provides a useful symmetric monoidal closed model for coherently commutative monoidal quasi-categories, complementing operadic and fibration-based models in the literature. The explicit Segal-condition description of fibrant objects and the proposed Quillen equivalence with symmetric monoidal quasi-categories are valuable and would give a concrete framework for comparing functorial and coCartesian-fibration models of symmetric monoidal higher categories. The paper also contains substantial auxiliary material, especially Appendix B on local objects in quasi-categorical enrichments, which is potentially useful independently. The main theorems are not machine-checked, but the constructions are explicit and mostly conventional, with no fitted parameters or ad hoc entities.","major_comments":[{"comment":"The proof relies on the assertion that if f: U → V is an (acyclic) cofibration in the strict JQ-model structure, then f ∗ Γ^n is again an (acyclic) cofibration for every n, but the text only says 'one can show' without proof. This preservation statement is the key step used to verify that Day convolution is a Quillen bifunctor on the strict model structure. The statement is in fact true by adjunction, since − ∗ Γ^n is left adjoint to Map(Γ^n, −) and precomposition with n+ ∧ − preserves Joyal fibrations and acyclic fibrations, but the proof must be written out. Without this step, Theorem 4.14 is not established, and the subsequent use of the strict monoidal model structure in Theorem 5.18 is unsupported.","section":"§4.1, proof of Theorem 4.14"},{"comment":"The proof that the pushout-product i□j is a JQ-equivalence when j is an acyclic JQ-cofibration invokes Proposition 5.13. That proposition applies only to a cofibration whose source and target are cofibrant objects in the model category. Here j: Y → Z is an arbitrary acyclic JQ-cofibration, and Y and Z are not assumed cofibrant. Moreover, the domain of i□j is the pushout U⊗Z ∐_{U⊗Y} V⊗Y, which is not known to be cofibrant because the monoidal model structure is exactly what is being proved. The subsequent closure of the class S under pushout, transfinite composition, and retract addresses only the first variable, and the 'same argument' for the second variable repeats the gap. A cellular reduction to generating acyclic cofibrations, or another direct verification of the pushout-product axiom, is needed before Theorem 5.18 can be accepted.","section":"§5, proof of Theorem 5.18"},{"comment":"Both theorems state that a fibration between two coherently commutative monoidal quasi-categories is a strict JQ-equivalence, and the proofs claim this 'follows from (1)'. This is false in general: in a left Bousfield localization, fibrations between fibrant objects need not be weak equivalences, and in any nontrivial model category the terminal map X → ∗ from a nonterminal fibrant object is a fibration between fibrant objects that is not a weak equivalence. Applied here, the claim would imply that every coherently commutative monoidal quasi-category is degreewise contractible. The ancillary assertion should be corrected or removed, and the proofs of Theorems 5.11 and 7.16 should not rely on it.","section":"§5, Theorem 5.11 and §7, Theorem 7.16"},{"comment":"The identification of the localized fibrant objects with symmetric monoidal quasi-categories depends on the categorical equivalence [N(k+/Γop), X♮]^♭_{Γop} ≃ U(X(k+)), which is quoted from [Sha, Lemma 3.12] without proof or even a precise statement. Since [Sha] is the author's earlier preprint, the reader cannot immediately verify this load-bearing step. Please include the statement of the lemma, a proof sketch, or the exact published reference; otherwise the final comparison with symmetric monoidal quasi-categories is conditional on an unverified external result.","section":"§8, Proposition 8.3 and Remark 8"}],"minor_comments":[{"comment":"The manuscript contains many typos and OCR-style artifacts, including 'cofibrations', 'immidiate', 'caontain', 'equilizer', 'coheretly', and inconsistent dashes. A careful proofreading pass is needed.","section":"Throughout"},{"comment":"The sentence 'It follows from [, Thm.] that the functor − ∗ Γn has a right adjoint' has an empty citation. Please provide the precise reference or a proof of the adjunction.","section":"§4.1, after Eq. (6)"},{"comment":"The adjoint pair is written as (N+•(Γop), F+•(Γop)), but the surrounding text and proof treat F+•(Γop) as the left adjoint. The order should be swapped for consistency, e.g., (F+•(Γop), N+•(Γop)).","section":"§8, Theorem 8.7"},{"comment":"The phrase 'equivalence of coheretly commutative monoidal categories' should read 'coherently commutative monoidal quasi-categories', and the same typo appears in several places.","section":"§5, Definition 5.9"}],"recommendation":"major_revision","confidential_remarks":"The central construction is plausible and likely repairable, but the manuscript in its current form has a genuine gap in the proof of the monoidal closure theorem (Theorem 5.18) and a false ancillary assertion in Theorems 5.11 and 7.16. I recommend major revision rather than rejection: the authors should supply the missing cellular argument, correct or remove the false statement, and clarify the status of the dependence on [Sha, Lemma 3.12]."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Bottom line: this is a real attempt to fill a known gap—a symmetric monoidal closed model category of coherently commutative monoidal quasi-categories—and the main architecture is probably right. The proof as written is not fully trustworthy, but the fixable parts are fixable.\n\nWhat is new: the JQ-model structure on Gamma-spaces, the symmetric monoidal closure under Day convolution, and the Quillen equivalences with normalized Gamma-spaces and with a localized model of coCartesian fibrations. This is a sensible extension of Schwede's machinery from the Kan to the Joyal model structure, and it addresses a gap that Kodjabachev--Sagave and others left open. That is genuine value.\n\nSome credit is due for the parts the stress-test flagged. The unproved 'one can show' preservation claim in Theorem 4.14 is actually true: - * Gamma^n is left adjoint to precomposition by n+ ∧ -, which preserves degreewise Joyal fibrations, so the preservation of (acyclic) cofibrations is immediate. The concern that Proposition 5.13 cannot apply in Theorem 5.18 because Y and Z are not cofibrant also dissolves: in the projective Joyal structure every object is cofibrant, since cofibrations are degreewise monomorphisms. The cellular reduction in Theorem 5.18 is terse, but the objection in the stress-test note does not land.\n\nThe real soft spot is Theorems 5.11 and 7.16. Both assert that a fibration between two fibrant objects is a strict JQ-equivalence. That is false in any model category: take the terminal map from a non-contractible coherently commutative monoidal quasi-category to the terminal Gamma-space. It is a fibration between fibrant objects but not a strict JQ-equivalence. The intended statement is presumably about acyclic fibrations, or some carefully phrased characterization of fibrations between fibrant objects in a left Bousfield localization. The proof of Theorem 5.18 uses a related but correct direction—strict JQ-fibrations between fibrant objects are fibrations in the localized model category—so the false ancillary assertions are not load-bearing for the main theorem, but they must be fixed.\n\nAlso worth noting: Section 8 leans on the author's earlier preprint [Sha] for Lemma 3.12, and Theorem 3.24 is cited from Lurie with a forward reference to Section 7. That is a real dependency on unpublished work, but not by itself disqualifying.\n\nWho should read this: people working on models for symmetric monoidal higher categories and Gamma-spaces. It deserves a serious referee, but the referee should require the fibration statements to be corrected and the proof of Theorem 5.18 written out more carefully before acceptance.","headline":"A useful model-categorical construction that likely works, but the written proof has a genuinely false assertion about fibrations and a couple of spots where the referee should demand more detail.","tokens_in":42653,"tokens_out":11493,"would_cite":false,"duration_ms":111899,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["18N60","18N55","18M05"],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper constructs a symmetric monoidal closed model category whose fibrant objects are coherently commutative monoidal quasi-categories, and proves it Quillen equivalent to a localized model category of symmetric monoidal…","keywords":["Gamma-spaces","model categories","quasi-categories","coherently commutative monoidal quasi-categories","Segal condition","Day convolution","left Bousfield localization","coCartesian fibrations"],"falsifier":"Find a single strict JQ-cofibration f : U → V and an n such that f ∗ Γ^n is not a JQ-cofibration; this can be checked on the generating cofibrations, since the claim is about the pushout-product axiom. Separately, the characterization of fibrant objects could be falsified by a Gamma-space satisfying the Segal condition whose fibrant replacement in the strict model category no longer satisfies it.","tokens_in":41610,"feed_emoji":"📐","tokens_out":9886,"duration_ms":89608,"temperature":0.7,"pith_summary":"The paper sets out to give symmetric monoidal higher categories a homotopy theory that is itself symmetric monoidal closed. Its central claim is that coherently commutative monoidal quasi-categories—Gamma-spaces satisfying the Segal condition—are exactly the fibrant objects of a newly defined JQ-model category on Gamma-spaces. It then constructs a companion model category on marked coCartesian fibrations over N(Γ^op), localizes it so that its fibrant objects are the standard symmetric monoidal quasi-categories, and proves the two model categories are Quillen equivalent. The payoff is that a common model for E-infinity structures—algebras over an E-infinity operad—lacked a closed monoidal model structure; Day convolution supplies the missing closure here.","feed_headline":"Coherently commutative monoidal quasi-categories get a model category","feed_subtitle":"A new model category for higher symmetric monoidal structures, closed under Day convolution.","key_machinery":"The central objects are Gamma-spaces, functors X : Γ^op → sSet, and the key mechanism is the set E∞S of localization maps h^k_l : Γ_k ⊔ Γ_l → Γ_{k+l}, whose local objects are exactly the Gamma-spaces satisfying the Segal condition. Dually, on the marked side the localization maps are Υ(k,l) between relative nerves N(k+/Γ^op) → N((k+l)+/Γ^op), and the relative nerve functor N^+_•(Γ^op) is the mediator of the final Quillen equivalence. The closed symmetric monoidal structure is carried by Day convolution, with the representable Gamma-space $Γ^{1}$ as unit, and the proof of monoidal closure reduces to a preservation property of the operation − ∗ Γ^n along cofibrations.","core_discovery":"On the category of Gamma-spaces (functors from Γ^op to simplicial sets), the paper defines the JQ-model category: a left Bousfield localization of the strict projective model structure obtained degreewise from the model structure for quasi-categories, with respect to the maps h^k_l : Γ_k ⊔ Γ_l → Γ_{k+l} induced by the two projections. The fibrant objects are precisely the Gamma-spaces satisfying the Segal condition X((k+l)+) → X(k+) × X(l+), and these are called coherently commutative monoidal quasi-categories. Theorem 5.11 establishes that these form a closed, left proper, combinatorial model category; Theorem 5.18 shows the closed symmetric monoidal structure given by Day convolution makes it a symmetric monoidal closed model category. Theorem 6.13 adds a Quillen equivalence with a normalized, strictly unital version, and Theorem 8.10 establishes a Quillen equivalence with the localized model category on marked coCartesian fibrations over N(Γ^op), whose fibrant objects are the standard symmetric monoidal quasi-categories.","pith_inferences":["This suggests a testable extension: running the same localization construction with the indexing category Γ^op replaced by a variant (e.g., for non-symmetric or braided structures) may yield closed monoidal model categories for other flavors of higher monoidal objects, provided the analogous preservation property holds.","This also suggests that the paper's comparison between the functor model and the coCartesian-fibration model likely lifts to an equivalence of the associated ∞-categories, not just a Quillen equivalence, which would make the two presentations interchangeable at the higher-categorical level.","If the unproved preservation assertion behind Theorem 4.14 is supplied, the same monoidal-closure argument would probably adapt to any cartesian closed model category playing the role of the quasi-category model structure, giving a general recipe for Gamma-object homotopy theories."],"forward_implications":["The homotopy category of coherently commutative monoidal quasi-categories is semi-additive: finite coproducts and products coincide up to homotopy, so the objects behave like a higher-categorical analogue of abelian monoids.","Every coherently commutative monoidal quasi-category can be rectified to a strictly unital one, because the normalization adjunction is a Quillen equivalence between the JQ-model category and the normalized JQ-model category.","Symmetric monoidal quasi-categories, presented as coCartesian fibrations over N(Γ^op), have a well-defined homotopy theory and are Quillen equivalent to the functor model, so results proved in either presentation carry over.","Mapping objects between a cofibrant Gamma-space and a coherently commutative monoidal quasi-category are again coherently commutative monoidal quasi-categories, so the internal function objects stay within the same homotopical world.","Because the model category is symmetric monoidal closed under Day convolution, one can form homotopy-coherent monoids and modules inside this model without leaving a symmetric monoidal closed setting."],"supporting_citations":[{"why":"Supplies the template for the strict and stable Q-model structures on Gamma-spaces that this paper generalizes to the quasi-categorical setting.","marker":"[Sch99]"},{"why":"Provides the marked simplicial set model structure, the coCartesian model structure on S+/N(Γ^op), and the relative nerve functor used in the final comparison.","marker":"[Lur09]"},{"why":"Proves the existence of left Bousfield localizations of combinatorial model categories, which the paper uses to construct the JQ-model categories.","marker":"[Bar07]"},{"why":"Introduces the convolution product on functor categories that provides the symmetric monoidal closed structure.","marker":"[Day70]"},{"why":"Constructs the smash product of Gamma-spaces that underlies the normalized JQ-model category.","marker":"[Lyd99]"},{"why":"The earlier lemma (Lemma 3.12) identifying mapping spaces into relative nerves, used to identify symmetric monoidal quasi-categories as local objects.","marker":"[Sha]"},{"why":"Provides the cartesian closed structure of the pointed simplicial set model used in the enrichment arguments.","marker":"[JT08]"}],"fun_headline_variants":["Symmetric monoidal closed model for coherent quasi-categories","Quillen equivalence for coherent monoidal quasi-categories","Day convolution gives closed model for coherent monoidal","Closed symmetric monoidal model for coherent Gamma-spaces"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The construction's monoidal closure rests on the unproved assertion, stated in the proof of Theorem 4.14 as 'one can show', that tensoring a (acyclic) cofibration of Gamma-spaces with the representable Γ^n yields another (acyclic) cofibration; if that assertion fails, the strict model category is not monoidal and the localized symmetric monoidal closure does not follow.","fun_headline_variants_meta":{"raw":{"variants":["Symmetric monoidal closed model for coherent quasi-categories","Quillen equivalence for coherent monoidal quasi-categories","Day convolution gives closed model for coherent monoidal","Closed symmetric monoidal model for coherent Gamma-spaces"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000796,"raw_usage":{"total_tokens":3441,"prompt_tokens":821,"completion_tokens":2620,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":437,"completion_tokens_details":{"reasoning_tokens":2568}},"tokens_in":437,"tokens_out":2620,"duration_ms":17836,"temperature":1.0,"reasoning_tokens":2568,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T13:17:04.989733+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Find a single strict JQ-cofibration f : U → V and an n such that f ∗ Γ^n is not a JQ-cofibration; this can be checked on the generating cofibrations, since the claim is about the pushout-product axiom. Separately, the characterization of fibrant objects could be falsified by a Gamma-space satisfying the Segal condition whose fibrant replacement in the strict model category no longer satisfies it.","supporting_citations":[],"review_version":1}