{"id":"447e8312-1e57-4924-950b-84ddf00f3e60","arxiv_id":"2506.01521","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Anomalous representations of a category with anomaly J are equivalent to Vect-linear functors on the extension C^J and to scalar representations of the Stolz-Teichner subcategory C^J_ST.","lead":"The authors prove a categorical theorem that turns anomalous or projective representations of a category into ordinary linear representations of two categories built from the anomaly. It generalizes the classical group-level link between projective representations and central extensions, and gives a conceptual home for anomaly-linearization tricks used in functorial quantum field theory.","discovery_kind":"unification","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The central triangle in Prop. 8.8 hinges on an unproved finite-dimensional Eilenberg–Watts step: fibers Mod_{J(X)} are not cocomplete, so 'cocontinuous' in Def. 6.9 is ill-typed, and Prop. 8.6 needs a finite-colimit Eilenberg–Watts theorem that is neither stated nor proved.","rationale":"The reader's weakest assumption and my independent check coincide. The paper is a serious, internally coherent formalization; the explicit constructions in §5–§7 are checkable and no circularity or fabrication appears. However, the central advertised result is Prop. 8.8, and the only step that imports substantial external input is the Eilenberg–Watts identification in Prop. 8.6 (and Lemma 8.3). The mismatch between the setting of the paper—finite-dimensional algebras, bimodules and vector spaces—and the usual Eilenberg–Watts hypotheses—module categories with all small colimits—is real and unaddressed. This is load-bearing because without that identification the inverse functor from C^J_ST-representations to anomalous representations is not shown to be essentially surjective on Hom_{Vect_K}(C^J,Vect_K); the triangle would not close. I do not think the flaw is necessarily fatal: the finite-dimensional version of Eilenberg–Watts is plausible and can be supplied, provided 'cocontinuous' is read as 'preserving finite colimits' and the lemma is proved. That is exactly the kind of repair that makes the paper's conclusion CONDITIONAL rather than fully established. The abstract and introduction overstate the result by dropping the additive/cocontinuous qualifier. No other concern of comparable weight emerged.","tokens_in":30675,"tokens_out":12700,"duration_ms":141877,"concrete_test":"Recompute Prop. 8.6 in the minimal case C = BG, G finite, with J = J_α for a nontrivial α. The unique fiber is Mod_K = Vect_K^fd. Isolate the assertion used: every additive K-linear functor E: Vect_K^fd -> Vect_K^fd preserving finite colimits is naturally isomorphic to (-⊗_K E(K)). Prove this assertion directly from the hypotheses of Def. 6.9, or exhibit an additive K-linear functor preserving finite direct sums and cokernels that is not of this form (e.g. by checking naturality of the maps E(K^n) ≅ E(K)^n for arbitrary bases). If a counterexample exists, Prop. 8.6 is false as stated; if the proof goes through, insert a lemma stating the finite-dimensional Eilenberg–Watts theorem used and replace 'small colimits' in Def. 6.9 by 'finite colimits'.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The proof of the right-hand equivalence in Prop. 8.8 is Prop. 8.6. There, for each object X of C, the restriction E|_X is treated as a cocontinuous functor Mod_{J(X)} -> Vect_K and Eilenberg–Watts is used to identify E|_X(L) with L ⊗_{J(X)} E|_X(J(X)). But by §2, Mod_{J(X)} is the category of finite-dimensional right J(X)-modules; when J(X)=K this is Vect_K^fd, which does not have small colimits (no infinite direct sums, no filtered colimits). Thus the hypothesis in Def. 6.9 that E is 'additive and cocontinuous (i.e., preserving small colimits) over C' is not well-formed for these fibers, and the classical Eilenberg–Watts theorem cited in §8.1 does not apply as stated. What the proof needs is a finite-dimensional variant: an additive K-linear functor preserving finite colimits (or at least cokernels) from finite-dimensional A-modules to Vect_K is naturally isomorphic to -⊗_A M. No such lemma is stated or proven. If the needed finite-colimit Eilenberg–Watts requires an extra hypothesis, or fails for some additive finite-cocontinuous functor, then the isomorphism in Prop. 8.6 does not follow and the triangle collapses; even if it holds, the abstract's unqualified 'Vect-linear functors' claim omits the essential restriction to additive/cocontinuous functors, making the headline statement stronger than what is established.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper develops a categorical framework for projective and anomalous representations. For a functor J:C→2Vect_K, it defines an anomalous representation as a lax homotopy commutative triangle relative to the terminal map, constructs the lax homotopy fiber C^J, and a Stolz–Teichner subcategory C^J_ST. The main result, Proposition 8.8, asserts a commuting triangle of equivalences among (a) anomalous representations with anomaly J, (b) Vect_K-linear additive cocontinuous functors C^J→Vect_K, and (c) linear representations of C^J_ST on which J acts as scalars. The group case with J_α recovers the classical equivalence between projective representations of class α and linear representations of the central extension G^α on which K^* acts as scalars, and the appendix recovers the twisted group algebra as a Kan extension.","tokens_in":31007,"tokens_out":9960,"duration_ms":105470,"significance":"If valid, Proposition 8.8 gives a universal linearization of anomalies, subsuming the Stolz–Teichner Clifford-linear construction and the classical theory of projective group representations. The paper's strengths are its explicit and detailed proofs, its careful unwindings of simplicial definitions, the verification of the group special case, and the absence of hidden fitting parameters. The main theorem is proved from categorical universal properties and checked against an external standard (Eilenberg–Watts); however, the key Eilenberg–Watts input is not established in the finite-dimensional setting used here, so the central claim is presently conditional on a missing lemma.","major_comments":[{"comment":"The hypothesis 'additive and cocontinuous (i.e., preserving small colimits) over C' in Definition 6.9 is not well-formed for the fiber categories used in the proof. By §2, Mod_{J(X)} is the category of finite-dimensional right J(X)-modules; for J(X)=K this is Vect_K^fd, which has no infinite direct sums or filtered colimits. Hence the restrictions E|_X of Remark 6.5 have no small colimits to preserve, and the classical Eilenberg–Watts theorem cited in §8.1 does not apply as stated. This matters because Lemma 8.3 and both parts of Proposition 8.6 use Eilenberg–Watts to identify E|_X with –⊗_{J(X)} E|_X(J(X)); Proposition 8.6 is exactly the right-hand equivalence in Proposition 8.8. Please replace 'small colimits' by 'finite colimits' (or otherwise make the finite-dimensional setting precise) and state and prove the needed finite-dimensional Eilenberg–Watts statement, including the version for natural transformations used in the second half of the proof of Proposition 8.6.","section":"§6, Definition 6.9; §8, Lemma 8.3 and Proposition 8.6"},{"comment":"The abstract's claim (i) and the introduction's corresponding sentence present anomalous representations as equivalent to 'Vect-linear functors E:C^J→Vect' without the additive and cocontinuous restriction. The body's Definition 6.9 defines Hom_{Vect_K}(C^J,Vect_K) only for functors that are additive and cocontinuous over C, and Proposition 6.10 lands in that subcategory. The unqualified headline is therefore stronger than what is proved and should be amended (e.g., 'additive and cocontinuous Vect-linear functors'), with the same correction made wherever the result is summarized.","section":"Abstract and §1"},{"comment":"Lemma 8.10 justifies the right Kan extension (8.6) by asserting that 'BG is small and 2Vect_K is complete'. Completeness of the Morita 2-category of finite-dimensional algebras, bimodules, and intertwiners is neither proved nor referenced, so the lemma as stated rests on an unverified input. For the group case the extension is constructed explicitly in Lemma 8.12 and Corollary 8.14 does not need the general completeness claim; please either give a proof or precise citation for completeness, or restrict Lemma 8.10 to the explicit construction for BG.","section":"Appendix, Lemma 8.10"}],"minor_comments":[{"comment":"There are numerous typographical slips, e.g., 'invesigate' in the abstract, 'abritray 8-category', 'subteltiles', 'construtions', and 'well kown'; also the symbol '8' appears in place of '∞' in several places (e.g., §1 and Remark 5.10). These should be corrected.","section":"Throughout"},{"comment":"The notation Hom_K(C^J_ST,Vect_K) is used before being explicitly defined; please insert a formal definition (the full subcategory of Hom(C^J_ST,Vect_K) on functors for which each λ_X is a K-algebra map).","section":"§8, after Lemma 8.3"},{"comment":"The proof writes the composite 'E_{Z_{E_ST}}' without first defining it; please introduce notation for the functor obtained by applying Proposition 6.10 to Z_{E_{ST}} and then restricting, so that the displayed comparisons are unambiguous.","section":"§8, Proposition 8.6"}],"recommendation":"major_revision","confidential_remarks":"The finite-dimensional Eilenberg–Watts gap is the only obstruction I see to accepting the main theorem; the needed lemma is standard and likely repairable, so I recommend major revision rather than rejection. The reliance on the thesis [Vup25] and on the forthcoming [Sch25] and [Vup26] for the general setting is acceptable, though those references are not yet publicly available; this is not a blocking issue."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Here's the short version: this paper has a genuine new claim—a triangle of equivalences linking anomalous representations of a category C with anomaly J, Vect_K-linear functors on the extension C^J, and representations of the Stolz–Teichner subcategory C^J_ST where J acts by scalars. The group case recovers the classical projective-representation/central-extension theorem, and the proof is unusually explicit, with diagrams unwound and the special case checked. That is real work, and it deserves credit.\n\nWhat is actually new is the construction itself: the extension C^J and the subcategory C^J_ST, and the proof that the three categories are equivalent. I did not find this triangle in the cited literature. The authors also honestly flag that the full infinity-categorical version is deferred to forthcoming work.\n\nThe soft spot is exactly where the stress-test note lands. Definition 6.9 defines Hom_{Vect_K}(C^J, Vect_K) as additive and cocontinuous functors 'over C', but the fibers Mod_{J(X)} are finite-dimensional module categories over a finite-dimensional algebra. They are not cocomplete—no infinite direct sums—so 'preserving small colimits' is not well-typed for these fibers. Then Prop 8.6 applies the classical Eilenberg–Watts theorem to these fibers to identify E|_X(L) with L ⊗_{J(X)} E|_X(J(X)). The cited theorem (Eil60, Wat60) is for categories of all modules, which are cocomplete. What the proof needs is a finite-dimensional variant: any additive K-linear functor preserving finite colimits (or cokernels) from Mod_A^fd to Vect_K is naturally isomorphic to -⊗_A M. No such lemma is stated or proved. So the proof of the right-hand equivalence in the triangle has a genuine gap, and the abstract's unqualified 'Vect-linear functors' overstates what is established; the additive/cocontinuous hypothesis is omitted.\n\nThat said, the flaw is likely repairable. The other two identities in the triangle—Prop 8.5 and Prop 8.7—do not use Eilenberg–Watts, and the group example checks out. If the authors add the finite-right-exact hypothesis to Def 6.9 and prove the finite-dimensional EW lemma, the triangle should stand.\n\nWho this is for: people working on anomalies in functorial field theory and higher representation theory. They will find the categorical scaffolding useful, but they should not take the abstract at face value until the EW leg is fixed. It deserves a serious referee, with a request to repair Def 6.9, prove the missing finite-dimensional Eilenberg–Watts statement, and soften the abstract.","headline":"Real new categorical triangle, but the Eilenberg–Watts leg is built on an ill-typed cocontinuity assumption and needs a finite-dimensional variant before the abstract can be taken at face value.","tokens_in":31548,"tokens_out":3970,"would_cite":true,"duration_ms":39851,"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":"The paper establishes a three-way equivalence making every anomaly-linearization of a category, including the fermionic field-theory construction, one universal categorical operation.","keywords":["anomalous representations","projective representations","lax homotopy pullback","2-vector spaces","cocontinuous functors","central extensions","anomaly linearization","∞-categories"],"falsifier":"Choose a finite group G and a nontrivial 2-cocycle $\\alpha$, and compute explicitly the two sides of the triangle for C=BG: the category of projective representations of G of class $\\alpha$ and the category of J_alpha-acts-as-scalars functors on (BG)^{J_alpha}_{ST}. A direct count of isomorphism classes, for example with the quaternion group and its nontrivial 2-cocycle, would either confirm the bijection or expose the missing cocontinuity hypothesis; any mismatch would falsify Proposition 8.8 as stated.","tokens_in":30452,"feed_emoji":"🔁","tokens_out":10683,"duration_ms":114426,"temperature":0.7,"pith_summary":"The paper's goal is to show that the standard fix for anomalous functorial field theories—extend the bordism category so that the anomaly becomes a linear action—is not ad hoc but forced by a universal categorical construction. Given any category C and any anomaly functor J from C into the 2-category of algebras, bimodules, and intertwiners, the paper builds an extension C^J of C and a smaller subcategory C^J_ST inside it. It then proves a three-way equivalence: anomalous representations of C with anomaly J are the same as Vect-linear, colimit-preserving functors out of C^J, and these are the same as ordinary linear functors out of C^J_ST on which J acts by scalars. This generalizes the classical correspondence between projective representations of a group and linear representations of its central extension, and it specializes to the bordism construction used to linearize anomalous conformal field theories.","feed_headline":"Anomalies in category representations all linearize the same way","feed_subtitle":"Any anomaly builds an extension whose linear functors are exactly the anomalous representations.","key_machinery":"The load-bearing object is the extension C^J, defined as the lax homotopy pullback of J against a point: an object is a pair (X,L_X) with L_X a right module over J(X), and a morphism is a morphism of C decorated by a compatible module homomorphism. Inside it sits the distinguished subcategory C^J_ST, the full subcategory on the objects (X,J(X)); since an algebra is a module over itself, a morphism there is a morphism of C equipped with a pointing, i.e., a chosen element of the bimodule J(f). The argument is carried by the recognition theorem for additive colimit-preserving functors between module categories—such a functor is necessarily tensoring with a module—applied fiberwise over each object of C. That theorem is what lets the paper pass from an anomalous representation Z to the functor L_X ↦ L_X ⊗_{J(X)} Z(X), and back, and it is also what isolates the 'J acts as scalars' condition as exactly the image of Vect_K-linear functors under restriction to C^J_ST.","core_discovery":"The central discovery is a commuting triangle of equivalences (Proposition 8.8) for any functor J:C→2Vect: the category of anomalous representations of C with anomaly J, the category of Vect_K-linear additive colimit-preserving functors out of C^J, and the category of linear functors on C^J_ST on which J acts as scalars are all equivalent. The extension C^J is the lax homotopy fiber of J, and C^J_ST is its full subcategory on the objects (X,J(X)), where J(X) is regarded as a right module over itself. Restriction from C^J to C^J_ST is an equivalence onto the 'J acts as scalars' functors, and the inverse is built by tensoring fibers with the representing module supplied by the additive-cocontinuous recognition theorem. In the single-object case C=BG with J=J_alpha, this triangle reduces to the classical equivalence between projective representations of G of class $\\alpha$ and linear representations of the central extension on which K^* acts by scalars; in the bordism case it reproduces the fermionic-anomaly linearization of conformal field theories.","pith_inferences":["A practical test the paper leaves implicit: to check whether a proposed linearization of an anomalous theory is complete, verify the 'J acts as scalars' condition on the ST subcategory rather than constructing the full extension by hand.","If the cocontinuity hypothesis proves too strong for finite-dimensional module categories, the equivalence likely survives in modified form for functors preserving only the colimits those categories actually have; the classical central-extension theorem for infinite-dimensional representations is a natural place to test this.","The same lax-pullback pattern suggests a hierarchy of higher anomalies: replacing Vect and 2Vect by n-vector spaces should linearize anomalous n-representations in the same way, with the extension playing the role of a higher central extension.","The right-extension reading of the twisted group algebra in the appendix hints that the whole construction can be rephrased as: the anomaly J determines a universal target algebra, and anomalous representations are modules over it; making that precise for general C would give an even more compact formulation of the triangle."],"forward_implications":["For every anomaly J, anomaly cancellation can be performed uniformly: replace C by C^J, and anomalous representations become ordinary Vect_K-linear functors.","The 'J acts as scalars' condition is a complete invariant: a linear functor on C^J_ST extends to C^J if and only if it satisfies it.","The classical projective-representation theorem is a special case: with C=BG and J the anomaly associated to a 2-cocycle alpha, the triangle is the correspondence with linear representations of the K^*-central extension on which K^* acts by scalars.","The fermionic and degree-n anomaly extensions of conformal bordism categories are recovered as special cases of the same triangle.","The same construction extends to super vector spaces and Hilbert-space targets, so the linearization mechanism is not specific to finite-dimensional vector spaces."],"supporting_citations":[{"why":"Supplies the theorem that additive colimit-preserving functors out of module categories are tensor products with a module, used throughout the proof of the triangle.","marker":"[Eil60]"},{"why":"Companion proof of the same additive-cocontinuous recognition theorem, cited together with [Eil60] wherever the paper applies it.","marker":"[Wat60]"},{"why":"Provides the bordism-category linearization construction that the paper recovers as a special case and that motivates the subcategory C^J_ST.","marker":"[ST04]"},{"why":"Provides the higher-categorical framework of nerves and lax homotopy pullbacks in which C^J and the main equivalences are formulated.","marker":"[Lur]"}],"fun_headline_variants":["Every category anomaly yields a linear representation extension","Anomalous category reps reduce to linear reps on extensions","One construction turns any category anomaly into linear reps","Category anomaly? Build an extension, get linear reps","Unified linearization for all category anomalies"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The equivalence rests on assuming the functors out of C^J preserve all colimits on each fiber category of finite-dimensional modules, a condition the paper does not prove is well defined there; if that fails, the triangle as stated is too strong.","fun_headline_variants_meta":{"raw":{"variants":["Every category anomaly yields a linear representation extension","Anomalous category reps reduce to linear reps on extensions","One construction turns any category anomaly into linear reps","Category anomaly? Build an extension, get linear reps","Unified linearization for all category anomalies"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000207,"raw_usage":{"total_tokens":1417,"prompt_tokens":977,"completion_tokens":440,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":593,"completion_tokens_details":{"reasoning_tokens":368}},"tokens_in":593,"tokens_out":440,"duration_ms":5233,"temperature":1.0,"reasoning_tokens":368,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T11:40:39.300685+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Choose a finite group G and a nontrivial 2-cocycle $\\alpha$, and compute explicitly the two sides of the triangle for C=BG: the category of projective representations of G of class $\\alpha$ and the category of J_alpha-acts-as-scalars functors on (BG)^{J_alpha}_{ST}. A direct count of isomorphism classes, for example with the quaternion group and its nontrivial 2-cocycle, would either confirm the bijection or expose the missing cocontinuity hypothesis; any mismatch would falsify Proposition 8.8 as stated.","supporting_citations":[],"review_version":1}