{"id":"a3a68ecc-b5c7-4199-a1e1-8e4c2f74c4c5","arxiv_id":"1908.09553","paper_version":1,"verdict":"ACCEPT","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Every C-homology theory on spaces over a small countable category C is isomorphic to the balanced smash product with a fixed C^op-spectrum, proved via a new external Spanier-Whitehead duality.","lead":"This paper proves a representation theorem for generalized homology theories of diagram spaces: for a countable indexing category, every such theory comes from a fixed spectrum over the opposite category. The proof introduces a new external Spanier-Whitehead duality between spectra over a category and spectra over its opposite, also used to construct Chern characters for rational theories.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"No significant objection identified: the main theorem is conditional on an explicit countability hypothesis and the proof chain is coherent.","rationale":"The reader's verdict of ACCEPT with moderate confidence seems appropriate. The countability assumption the reader identifies as the weakest premise is indeed the principal hypothesis, but the paper states it explicitly in Theorem 5.2.3 and proves in Proposition 5.3.1 that it is equivalent to the more familiar condition that C is countable. Thus it is a limitation of scope, not a hidden fragility in the argument. The proof of the representation theorem is a coherent application of external duality and Neeman's representability theorems; the extension from finite C-CW-spectra to arbitrary C-spaces uses standard colimit and CW-approximation arguments. The only residual point is the slightly strengthened citation of Neeman's Proposition 4.11, which the author says has the same proof. I do not see evidence that this strengthening fails, and it can be checked directly against the cited source. I therefore see no reason to move the verdict.","tokens_in":39771,"tokens_out":36972,"duration_ms":382263,"concrete_test":"Verify Theorem 5.2.6 against Neeman's original Proposition 4.11: check whether the hypotheses permit an arbitrary essentially small triangulated subcategory S of compact generators, or require S = T^c. If the latter, verify that S_W^{C^op} coincides with the full subcategory of compact objects in S_H^{C^op}; this should follow since S_W^{C^op} is the triangulated closure of the representables and finite C-CW-spectra are compact. If the coincidence fails, re-run the proof of Theorem 5.2.6 with S_W^{C^op} in place of T^c to confirm the strengthening.","verdict_should_be":"UNCHANGED","load_bearing_attack":"After reading the argument in good faith, I do not find a load-bearing flaw in the central claim. Theorem 5.2.3 is stated with the explicit hypothesis that S_W^{C^op} is countable, and Proposition 5.3.1 shows this is equivalent to countability of C, so the theorem does not overclaim. The proof uses external Spanier-Whitehead duality to convert the given homology theory into a homological functor on finite C^op-spectra, then invokes Neeman's Brown representability results. The duality functor D is exact on triangulated categories (Lemma 4.2.5), finite C-CW-spectra are dualisable (Corollary 4.2.7), and the resulting representing C^op-spectrum is then extended from finite complexes to all C-spaces by the colimit argument. The one technical point that deserves a check is Theorem 5.2.6: the paper states a version of Neeman's Proposition 4.11 for an essentially small triangulated subcategory S of a triangulated category T, where Neeman's original formulation may ask for S to be the full subcategory of compact objects. The author asserts that the same proof applies. This is a citation-level strengthening rather than a detected mathematical error, and it is plausible because S_W^{C^op} is the natural compact generating subcategory of S_H^{C^op}. I therefore do not treat it as a reason to reject or conditionally accept the paper.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper develops an external Spanier-Whitehead duality for diagram spectra. For a small category C satisfying a cofibrancy condition, it constructs a duality functor D: (S_W^C)^op -> S_W^{C^op} as part of a closed bicategory of derived bimodules, and proves that every finite C-CW-spectrum is dualisable (Cor. 4.2.7). The main homological Brown representability theorem (Thm. 5.2.3) states that if S_W^{C^op} is countable (equivalently, if C is countable, by Prop. 5.3.1), every C-homology theory is naturally isomorphic to h^C_*(-; E) for some C^op-spectrum E, and every morphism of homology theories is induced by a morphism in the derived category. The proof converts the given homology theory into a homological functor on finite C^op-spectra via the duality functor and invokes Neeman's representability theorems. The final section passes to rational spectra and chain complexes through the stable Dold-Kan correspondence, proving existence of Chern characters for rational C-homology theories with flat coefficients (Cor. 6.3.7) and for orbit categories of finite groups with cyclic prime-power subgroups (Thm. D), with a characterisation in terms of hereditary category algebras (Prop. 6.5.1).","tokens_in":40048,"tokens_out":22985,"duration_ms":230997,"significance":"If the results stand, the paper completes the homological version of Brown representability for diagram spaces, complementing the easier cohomological case and showing that Davis-Lueck type constructions exhaust all C-homology theories under a mild countability hypothesis. The external duality functor is a genuine contribution, and the rational applications to Chern characters are new. The paper is honest about its hypotheses: Theorem 5.2.3 carries an explicit countability assumption, the rational flatness theorem is conditional, and the author openly indicates where proofs are sketched. The derivation is a traditional mathematical proof built on established external theorems (Neeman, Shulman, Shipley, Li, Thevenaz-Webb); there are no fitted parameters or empirical components. The careful statement of hypotheses and the explicit use of external results are strengths, and I found no load-bearing flaw in the central argument.","major_comments":[],"minor_comments":[{"comment":"The preamble to Theorem 5.2.6 states a version of [Nee97, Prop. 4.11] in which an essentially small triangulated subcategory S of compact objects generating T replaces Neeman's T^c, and says that 'the same proofs apply'. Since this theorem is used to produce the representing C^op-spectrum, please add a sentence (or a pointer to the exact passage in [Nee97]) explaining how Neeman's proof adapts to this slightly more general S. I do not regard the issue as fatal, but as written the reader has to reconstruct the argument.","section":"Theorem 5.2.6 (preamble)"},{"comment":"The transition 'From now on, C is a discrete index category' is abrupt: the earlier machinery was developed for SpO-enriched categories satisfying (C), while Theorem 5.2.3 is for ordinary categories. Please state explicitly that a Set-category is regarded as an SpO-category via the free enrichment (mapping spectrum Sigma^infty_+ Hom_C(c,d)), and note that this satisfies (C).","section":"Section 5, beginning"},{"comment":"The verification that the shuffle map nabla is a map of symmetric spectra is compressed into 'an easy diagrammatic check'. Since this compatibility is needed for the comparison Phi(HQ wedge Sigma^infty X) congruent N~QX underlying the rational Chern characters, please expand the check or give a precise reference.","section":"Proposition 6.1.1"},{"comment":"The h-cofibration induction in the proof of Proposition 2.3.2 is quite terse, especially the preservation of h-cofibrations under balanced smash products and the passage to limit ordinals. A few more details would improve verifiability, although the argument appears sound.","section":"Proposition 2.3.2"},{"comment":"There are minor typographical and grammatical issues, including 'developped' in Section 1, 'The proof are technically' in Section 2.3, and inconsistent notation for the derived category of C^op-spectra in Theorem 5.2.3. These do not affect the mathematics.","section":"Throughout"}],"recommendation":"minor_revision","confidential_remarks":"The paper is well within the scope of math.KT. The most delicate point is the use of Neeman's theorem in the stated generality; I recommend asking for a one-sentence justification but do not see a substantive flaw. Self-citation to [Lac16] is used for standard equivalences and is not load-bearing."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The short version: this is a solid paper with a genuinely new construction and a real theorem. The homology representation theorem for diagram spaces (Theorem 5.2.3) is new as far as I know, and the external Spanier-Whitehead duality functor D: finite C-spectra -> finite C^op-spectra is the right tool for it. The countability hypothesis is explicit and shown equivalent to C being countable, so the paper does not overclaim.\n\nWhat the paper does well: it sets up a closed bicategory of derived bimodules and proves external duality in that framework; it carefully compares different model categories of spectra, including the subtle weak monoidal Quillen equivalences needed for the rational case; it gives a clean countability argument; and the rational section provides useful Chern character criteria, including the hereditary category algebra characterization and the UFP result for orbit categories. The proof of Theorem 5.2.3 is coherent: dualize the homology theory to a cohomological functor on finite C^op-spectra, represent via Neeman, then extend by colimits. I did not find circular reasoning or fitted parameters.\n\nThe main thing I would want a referee to check is the use of Neeman's Proposition 4.11 in Theorem 5.2.6. The author states a slightly stronger version than Neeman's original formulation: essentially small triangulated subcategory S generating T and consisting of compact objects, rather than S being the full subcategory of all compact objects. The author says 'the same proofs apply.' That is plausible because S_W^{C^op} is the natural compact generating subcategory, but it is a strengthening of the cited result and should be verified carefully. Also, some technical lemmas are only sketched: the h-cofibration argument in Prop 2.3.2 and the shuffle map verification in Prop 6.1.1. These look routine but intricate; a referee should ask for more detail. None of this is a load-bearing flaw in my reading; the central argument holds up.\n\nThis paper is for people working on equivariant stable homotopy theory, assembly maps, and diagram spectra. It resolves an open question and provides a reusable framework. It deserves a serious referee; I would accept it with the expectation that the Neeman citation check and the technical lemmas get attention.","headline":"A genuinely new external duality functor and a clean homology representation theorem for diagram spaces, with a countable-category hypothesis that is explicit and shown necessary; the main soft spot is a citation-level strengthening of Neeman's theorem that should be checked.","tokens_in":40554,"tokens_out":1745,"would_cite":true,"duration_ms":18670,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["55N91","55M05","55P42","18D05"],"pacs":[],"model":"deepseek-v4-flash","headline":"The paper proves that every C-homology theory for a countable diagram category is a balanced smash product with a C^op-spectrum, and that maps of theories are maps of spectra.","keywords":["diagram spaces","C-homology theories","external Spanier-Whitehead duality","Brown representability","Chern character","orbit category","category algebras","Mackey functors"],"falsifier":"Work out the arrow category $C=(0\\to 1)$ explicitly: enumerate the finite $C$-CW-spectra and test every homological functor on $S_W^C$ for representability as $\\pi_*(-\\wedge_C E)$ with $E$ a $C^{\\mathrm{op}}$-spectrum. The theorem predicts all are representable, so even one homological functor that fails would refute Theorem 5.2.3.","tokens_in":39581,"feed_emoji":"📐","tokens_out":11565,"duration_ms":111887,"temperature":0.7,"pith_summary":"Diagram spaces are functors from a small category $C$ to pointed spaces, and $C$-homology theories are the natural homology theories for them. This paper proves a homological Brown representability theorem: whenever $C$ is countable, every $C$-homology theory $h^C_*$ is naturally isomorphic to $h^C_*(-;E)=\\pi_*(E\\wedge_C -)$ for some fixed $C^{\\mathrm{op}}$-spectrum $E$, and every morphism between such theories is induced by a morphism of spectra. The proof works by building an external Spanier-Whitehead duality that relates finite $C$-spectra to finite $C^{\\mathrm{op}}$-spectra, converting homology into cohomology where representability is easier. As applications, the paper constructs Chern characters for rational $C$-homology theories whose coefficient systems are flat, and for all rational theories over orbit categories of finite groups whose subgroups are cyclic of prime power order. The result matters because the smash-product construction underlies most equivariant and assembly-map homology theories, so the theorem says this construction is not missing any homology theory.","feed_headline":"Every homology theory on diagram spaces has a spectrum representative","feed_subtitle":"A new external duality converts homology to cohomology, yielding representatives and Chern characters.","key_machinery":"The load-bearing object is the external Spanier-Whitehead duality functor $D_{A,B}(X)=\\mathrm{Rmap}_A(X,A)$, defined on the closed bicategory $\\mathrm{DerMod}(\\mathrm{Sp}_O)$ whose 1-morphisms are derived bimodules between spectrally enriched categories. Where classical duality pairs a spectrum with another spectrum in the same category, this external version sends an $(A,B)$-bimodule to a $(B,A)$-bimodule, so it pairs $C$-spectra with $C^{\\mathrm{op}}$-spectra. For finite $(A,B)$-CW-spectra it is a true duality with $DDX\\cong X$, yielding an equivalence between the Spanier-Whitehead categories $S_W^C$ and $(S_W^{C^{\\mathrm{op}}})^{\\mathrm{op}}$. This is what lets the paper trade a homology theory on $C$-spectra for a representable cohomology theory on $C^{\\mathrm{op}}$-spectra.","core_discovery":"The central discovery, Theorem 5.2.3, is that homological Brown representability holds for diagram spaces: with $S_W^{C^{\\mathrm{op}}}$ countable (equivalently, with $C$ countable), every $C$-homology theory $h^C_*$ is naturally isomorphic to $h^C_*(-;E)$ for a $C^{\\mathrm{op}}$-spectrum $E$, and every natural transformation $h^C_*(-;E)\\to h^C_*(-;E')$ is induced by a morphism $E\\to E'$ in the derived category of $C^{\\mathrm{op}}$-spectra. This completes the representability picture for diagram spectra, since the cohomological version was already known. The proof passes through an external Spanier-Whitehead duality functor $D$ that pairs finite $C$-spectra with finite $C^{\\mathrm{op}}$-spectra, uses $D$ to convert the given homology functor into a cohomology functor on the opposite Spanier-Whitehead category, and then invokes the Brown representability results from Section 5.2 to obtain the representing $C^{\\mathrm{op}}$-spectrum.","pith_inferences":["The same external duality could plausibly prove a representation theorem for equivariant homology theories (the paper's Question 1), since groupoid spectra are again diagram spectra and the countability check would reduce to countability of the groupoid category.","The countability hypothesis is probably not removable: the Brown representability theorem used here genuinely requires countable triangulated categories, so uncountable $C$ is the most promising place to look for an unrepresentable $C$-homology theory.","Checking the unique factorisation property gives a purely combinatorial way to decide whether a finite EI category has universal Chern characters; this could be applied to concrete orbit categories without computing any spectra.","Extending the flat-coefficient Chern character theorem to infinite groups would need a flatness criterion weaker than the Mackey algebra's semisimplicity; the paper's $D_\\infty$ example shows von Neumann regularity already fails, pointing toward a homological rather than ring-theoretic condition."],"forward_implications":["The Davis-Lück construction $h^C_*(-;E)$ is complete for countable $C$: every $C$-homology theory has a representing $C^{\\mathrm{op}}$-spectrum.","External duality gives an equivalence of triangulated categories $S_W^C\\cong (S_W^{C^{\\mathrm{op}}})^{\\mathrm{op}}$, so duality interchanges the roles of $C$ and $C^{\\mathrm{op}}$ in all homological statements.","Rational $C$-homology theories with flat coefficient systems decompose into Bredon homology summands through a Chern character.","For finite orbit categories $\\mathrm{Or}(G,F)$ with $G$ finite and all members of $F$ cyclic of prime power order, every rational $C$-homology and cohomology theory has a Chern character.","For finite EI categories, the existence of Chern characters for all rational theories is equivalent to the category algebra $QC$ being hereditary, equivalently to the unique factorisation property."],"supporting_citations":[{"why":"Supplies the Brown representability theorems (5.2.5 and 5.2.6) for countable triangulated categories that produce the representing C^op-spectrum.","marker":"[Nee97]"},{"why":"Supplies the closed bicategory duality theory from Chapter 16 that the external duality construction is modelled on.","marker":"[MS06]"},{"why":"Introduces the construction h^C_n(X;E)=pi_n(E wedge_C X) whose converse is the paper's main theorem.","marker":"[DL98]"},{"why":"Establishes the projective model structure on enriched functor categories under condition (C), setting up C-spectra.","marker":"[Shu06]"},{"why":"Provides the model category of orthogonal spectra and the monoid and pushout-product axioms the whole framework uses.","marker":"[MMSS01]"},{"why":"Gives the stable Dold-Kan correspondence as a zig-zag of weak monoidal Quillen equivalences, used for the rational case.","marker":"[Shi07]"},{"why":"Classifies hereditary category algebras of finite EI categories, needed for the Chern character criterion in Section 6.5.","marker":"[Li11]"},{"why":"Provides the Mackey algebra basis and semisimplicity that yield flatness for coefficient systems extending to Mackey functors.","marker":"[TW95]"}],"fun_headline_variants":["Diagram homology theories now have spectrum reps via duality","External duality proves homology representability on diagrams","Homological Brown representability achieved for diagram spaces","Spectrum representatives for all diagram homology theories","Duality converts homology to cohomology on diagram spectra"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The proof's load-bearing premise is that the indexing category $C$ is countable: the Brown representability theorems it cites require the Spanier-Whitehead category $S_W^{C^{\\mathrm{op}}}$ to be countable, and Proposition 5.3.1 proves this happens exactly when $C$ is countable.","fun_headline_variants_meta":{"raw":{"variants":["Diagram homology theories now have spectrum reps via duality","External duality proves homology representability on diagrams","Homological Brown representability achieved for diagram spaces","Spectrum representatives for all diagram homology theories","Duality converts homology to cohomology on diagram spectra"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000211,"raw_usage":{"total_tokens":1368,"prompt_tokens":856,"completion_tokens":512,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":472,"completion_tokens_details":{"reasoning_tokens":441}},"tokens_in":472,"tokens_out":512,"duration_ms":5247,"temperature":1.0,"reasoning_tokens":441,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T11:07:54.403658+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Work out the arrow category $C=(0\\to 1)$ explicitly: enumerate the finite $C$-CW-spectra and test every homological functor on $S_W^C$ for representability as $\\pi_*(-\\wedge_C E)$ with $E$ a $C^{\\mathrm{op}}$-spectrum. The theorem predicts all are representable, so even one homological functor that fails would refute Theorem 5.2.3.","supporting_citations":[],"review_version":1}