{"id":"e58b18bc-1ad9-44fa-a465-0cdb3cacb427","arxiv_id":"2411.08746","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"The inclusion of an exact form category into its bounded chain complexes induces a homotopy equivalence of Grothendieck-Witt spaces for the new quadratic functor (Theorem 10.5).","lead":"This mathematics paper proves that a certain invariant, Hermitian K-theory, does not change when you move from an exact category to its richer category of bounded chain complexes, using a newly defined notion of quadratic forms on chain complexes. This is groundwork for reconciling the classical and modern (infinity-categorical) formulations of Hermitian K-theory.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Lemma 10.6's reduction to semi-idempotent completion is cited from a Cofinality theorem that only covers isomorphism weak equivalences; as written, the w-version needed for Theorem 10.5 is unproved.","rationale":"The reader identified Lemma 10.6 and the semi-idempotent-completion reduction as the weakest assumption, and my stress-test converges on the same point. The specific aggravating detail is that the proof of Lemma 10.6 cites Theorem 4.2 and Example 4.3, both of which concern Grothendieck-Witt spaces with isomorphisms as weak equivalences, while the lemma is stated for arbitrary weak equivalences w. This is not an internal inconsistency in the sense of a false statement, and the theorem may well be true; however, as written, the proof does not justify the first reduction on which Theorem 10.5 is built. The remainder of the argument depends on that reduction in a concrete way: strict acyclicity in Lemma 10.4 is used to obtain admissible monomorphisms with specific kernels, and the filtration in diagram (10.2) is only available after passing to a semi-idempotent complete category. Since the reader already returned a CONDITIONAL verdict, my finding does not change the verdict; it sharpens the condition that should be attached: prove or explicitly cite a cofinality theorem for GW(-,w,-) valid for arbitrary weak equivalences, and verify that acyclic bounded complexes in a semi-idempotent complete exact category are strictly acyclic in the sense needed by Lemma 10.4.","tokens_in":57695,"tokens_out":13446,"duration_ms":121136,"concrete_test":"Re-derive Lemma 10.6 from the resources of the paper: try to instantiate the proof of Theorem 9.8 with the inclusion E subset E~0, or check the proof of the cited [Sch10b, Lemma 12] to see whether it covers GW(-,w,-) for arbitrary weak equivalences without a symmetric-cone hypothesis. If no such proof exists without a strong symmetric cone, then Theorem 10.5 currently rests on an unproved reduction and must be marked CONDITIONAL until the missing w-version of the cofinality statement is supplied; if [Sch10b, Lemma 12] does prove exactly this statement, the reduction is sound and the concern does not land.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Theorem 10.5 depends at its first step on Lemma 10.6, which asserts that passing to the semi-idempotent completion E~0 does not change GW(E,w,Q). This reduction is used in two essential places: Lemma 10.4 uses the consequence that acyclic complexes in Ch^b E are strictly acyclic, from which the admissible monomorphism [d_n; phi_n] is derived, and the homotopy-cartesian diagram (10.2) in the proof of Theorem 10.5 is built only after this reduction. However, the proof of Lemma 10.6 cites Theorem 4.2 and Example 4.3, which are statements about GW(E,Q) for exact form categories with isomorphisms as weak equivalences. For a general exact form category with weak equivalences (E,w,#,can,Q), the space GW(E,w,Q) is defined using wQuad(R_bullet E,Q) and wS_bullet E, and it is not a special case of the Part 1 Grothendieck-Witt space. The paper does not state a cofinality theorem for GW(-,w,-) without a strong symmetric cone; Theorem 9.8, the only Part 2 cofinality theorem, explicitly assumes a strong symmetric cone, which is not assumed in Lemma 10.6. Thus the reduction to semi-idempotent complete categories is not established by the citations given. This is a proof gap rather than a demonstrated counterexample: if [Sch10b, Lemma 12] indeed proves the analogous statement for arbitrary weak equivalences, the concern is resolved, but the present text does not supply that argument.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper is the second installment of the author's programme on higher Hermitian K-theory. It develops the theory of exact form categories with weak equivalences and establishes the structural theorems Additivity (Theorem 8.1), Fibration (Theorem 9.3), Cofinality (Theorem 9.8), and an algebraic Bott sequence (Theorem 11.9). Its main new result is Theorem 10.5: for any exact form category with strong duality (E, sharp, can, Q), the inclusion of E into bounded chain complexes Ch^b E as degree-zero complexes induces a homotopy equivalence GW(E,Q) -> GW(Ch^b E, quis, Q). The proof constructs an explicit quadratic functor on chain complexes (Definition 10.2), proves it is quadratic left exact with a strong symmetric cone (Lemma 10.3), reduces to semi-idempotent complete categories via Lemma 10.6, and combines Additivity and Fibration with a GW0-surjectivity argument (Lemma 10.4).","tokens_in":58135,"tokens_out":9492,"duration_ms":77521,"significance":"If the results are correct, Theorem 10.5 is a substantial bridge: it shows that the classical 1-categorical Hermitian K-theory of exact categories is invariant under passage to bounded chain complexes with quasi-isomorphisms, a key step for matching with the infinity-categorical treatment of Calmes-Dotto-Harpaz-Hebestreit-Land-Moi-Nardin-Nikolaus-Steimle. The paper also provides a clean framework of Grothendieck-Witt spectra and Bott sequences for exact form categories with weak equivalences. The construction of the quadratic functor on Ch^b E is explicit and checkable, and several proofs, such as Lemma 10.3 and the presentation of GW0 in Theorem 8.4, are written in detail. The main caveat is that a load-bearing reduction to semi-idempotent complete categories is only cited, not proved in the present text.","major_comments":[{"comment":"The semi-idempotent completion reduction is not established by the citations given. Lemma 10.6 is stated for GW(E,w,Q) with an arbitrary set of weak equivalences, but its proof invokes Theorem 4.2 and Example 4.3, both of which concern Part 1 Grothendieck-Witt spaces GW(E,Q) with isomorphisms as the weak equivalences. The only cofinality result in Part 2, Theorem 9.8, assumes a strong symmetric cone, and Lemma 10.6 has no such hypothesis. Since the proof of Theorem 10.5 starts with 'By Lemma 10.6 below, we can assume E is semi-idempotent complete', this is a load-bearing gap: the step from E to E~0 changes the category in which the chain-complex acyclicity and strict acyclicity arguments are run. The author should either prove a cofinality statement for GW(-,w,-) without a symmetric cone, or reproduce or quote the precise statement and proof of [Sch10b, Lemma 12] and explain how Theorem 4.2 and Example 4.3 apply in the w-setting.","section":"Section 10, Lemma 10.6"},{"comment":"The proof of Theorem 10.5 asserts, immediately after invoking Lemma 10.6, that in a semi-idempotent complete exact category acyclic bounded chain complexes are strictly acyclic; Lemma 10.4 uses this to identify the map [d_n; phi_n] as an admissible monomorphism. The implication is not proved and no precise reference is given. If this is a known consequence of semi-idempotent completeness, please provide a citation or a short argument; otherwise the surjectivity proof of GW0 and the construction of the admissible monomorphisms in Lemma 10.4 are incomplete.","section":"Section 10, Theorem 10.5 and Lemma 10.4"}],"minor_comments":[{"comment":"The word 'spectrum' is misspelled as 'spetrum' in the definition of the Grothendieck-Witt spectrum.","section":"Definition 11.5"},{"comment":"There are minor typos: 'contranctible' should be 'contractible' and 'surjectve' should be 'surjective'.","section":"Proof of Theorem 10.5"},{"comment":"The abstract contains a spacing typo: 'W e prove' should be 'We prove'.","section":"Abstract"},{"comment":"The phrase 'the obvious one’s' should be 'the obvious ones'.","section":"Lemma 2.10"},{"comment":"Several statements are delegated to previous papers with 'mutatis mutandis' (for example Proposition 2.5, Theorem 3.2, Corollary 3.3, Lemma 5.6, Lemma 8.3, and parts of Proposition 9.6). For a sequel this is acceptable, but the paper would be easier to verify if each such passage indicated which quadratic-form inputs need to be checked, especially in Lemma 8.3 and Proposition 9.6.","section":"Various sections"}],"recommendation":"major_revision","confidential_remarks":"The main theorem is plausible and the overall strategy is coherent, but the proof of Lemma 10.6 is a genuine gap in the written argument. If the author can supply a correct cofinality argument for GW(-,w,-) without a symmetric cone, or a precise reference and proof sketch for the cited statement, I expect the paper to be publishable."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"First thing you should know: this is the paper that makes good on the promise to extend the quadratic functor from an exact form category to bounded chain complexes without changing Grothendieck-Witt spaces (Theorem 10.5). That is genuinely new for arbitrary quadratic forms with values in abelian groups; the symmetric-form case was [Sch10b, Prop 6]. The definition of Q on chain complexes (Definition 10.2) is the main contribution, and the proof of Theorem 10.5 via the filtration by supports and the Fibration/Additivity theorems is a real piece of work. The paper also proves useful general results: Additivity, Fibration, Cofinality for form categories with weak equivalences, and constructs the Grothendieck-Witt spectrum and Bott sequence (Theorems 8.1, 9.3, 9.8, 11.9). These are stated for the quadratic setting and the adaptions are mostly clean.\n\nThe soft spots are real but not disqualifying. First, the paper delegates a large fraction of proofs to earlier work with 'mutatis mutandis' (Prop 2.5, Thm 3.2, Cor 3.3, Lemmas 5.6, 8.3, parts of Prop 9.6). That is normal for this author and this series, but it makes independent verification slow. Second, and more concretely, Lemma 10.6 asserts that passing to the semi-idempotent completion does not change GW(E,w,Q), and the proof cites Theorem 4.2 and Example 4.3, which are cofinality statements for exact form categories with isomorphisms as weak equivalences. The w-version of cofinality is not supplied; Theorem 9.8 has a strong symmetric cone hypothesis not assumed in Lemma 10.6. As written, the reduction to semi-idempotent complete categories is not established by the citations. If [Sch10b, Lemma 12] does prove the w-version, a reference with the argument appended fixes it; the present text does not give that argument. This is a proof gap, not a demonstrated counterexample, and it is load-bearing for Theorem 10.5 because the proof uses strict acyclicity of acyclic complexes after the reduction.\n\nThe citation pattern is heavily self-referential, but that is appropriate for a series where the prior papers exist and the results are real. There is no code or machine-checked formalization to lean on; this is classical proof-based mathematics, and the reader has to trust the \"mutatis mutandis\" transfers.\n\nWho is this for? People working in Hermitian K-theory, quadratic forms, and the comparison with the infinity-categorical setup of CDH+. It deserves a serious referee: the main theorem is important and the overall structure is sound, but the referee should demand the Lemma 10.6 repair and probably fuller proofs of the delegated quadratic statements. I would engage with it; it is the right kind of paper to send to a careful referee.","headline":"Foundational second paper in Schlichting's Hermitian K-theory series; genuinely new quadratic-form extension to chain complexes, but the reduction to semi-idempotent completion has a citation gap that needs checking.","tokens_in":58596,"tokens_out":2566,"would_cite":true,"duration_ms":22252,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["19G38","19D06"],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper proves that for any exact form category with strong duality, the Grothendieck-Witt space is homotopy equivalent to the Grothendieck-Witt space of its bounded chain complexes with quasi-isomorphisms.","keywords":["Hermitian K-theory","Grothendieck-Witt groups","quadratic forms","exact categories","bounded chain complexes","quasi-isomorphisms","Bott sequence","cofinality"],"falsifier":"Check the surjectivity step of Lemma 10.4 on an exact category that is not semi-idempotent complete, for instance finitely generated free modules over a ring possessing a non-free stably free module. If some nondegenerate quadratic form on a bounded complex cannot be rewritten as a degree-zero form plus hyperbolics, then $GW_0(E,i,Q) \\to GW_0(\\mathrm{Ch}^b E, \\mathrm{quis}, Q)$ is not surjective and Theorem 10.5 is false for that category.","tokens_in":57507,"feed_emoji":"🔗","tokens_out":11303,"duration_ms":93098,"temperature":0.7,"pith_summary":"This paper aims to put Hermitian K-theory of exact categories on the same footing as algebraic K-theory by moving the quadratic data to chain complexes. It proves that the inclusion of an exact form category with strong duality $E$ into bounded chain complexes concentrated in degree zero induces a homotopy equivalence of Grothendieck-Witt spaces $GW(E,i,Q) \\simeq GW(\\mathrm{Ch}^b E, \\mathrm{quis}, Q)$. The vehicle is a new extension of the quadratic functor $Q$ to all bounded complexes, defined by forms that are compatible with the differential. This gives Hermitian K-theory the standard chain-complex machinery: additivity, fibration, cofinality, and a delooped spectrum with a Bott sequence.","feed_headline":"Chain complexes don't change Grothendieck-Witt groups","feed_subtitle":"A new quadratic functor on bounded complexes equips Hermitian K-theory with chain-level tools and a Bott sequence.","key_machinery":"The load-bearing object is the quadratic functor $Q$ on bounded chain complexes introduced in Definition 10.2. A form on a complex is a pair $(\\xi,\\varphi)$ consisting of a symmetric chain map $\\varphi \\colon E \\to E^\\sharp$ and a quadratic form $\\xi$ on the degree-zero object, linked by $d_1^\\bullet(\\xi)=0$ and $\\rho(\\xi)=\\varphi_0$. This functor is quadratic left exact, so $\\mathrm{Ch}^b E$ becomes an exact form category with quasi-isomorphisms as weak equivalences; the paper also uses a strong symmetric cone, a cone/path-object pair that detects quasi-isomorphisms through acyclicity, to run the fibration and Bott-sequence arguments.","core_discovery":"The central claim is Theorem 10.5: for any exact form category with strong duality, the degree-zero inclusion $E \\subset \\mathrm{Ch}^b E$ is a homotopy equivalence of Grothendieck-Witt spaces, $GW(E,i,Q) \\simeq GW(\\mathrm{Ch}^b E, \\mathrm{quis}, Q)$. The proof works by defining, on a bounded complex $(E,d)$, a quadratic form to be a pair $(\\xi,\\varphi)$ with $\\varphi \\colon E \\to E^\\sharp$ symmetric, $\\xi \\in Q(E_0)$, $d_1^\\bullet(\\xi)=0$, and $\\rho(\\xi)=\\varphi_0$. This makes $\\mathrm{Ch}^b E$ into an exact form category with weak equivalences and a strong symmetric cone. The paper further establishes additivity, fibration and cofinality theorems for exact form categories with weak equivalences, and constructs a Grothendieck-Witt spectrum whose negative homotopy groups are Witt groups, with an algebraic Bott sequence $GW^{[n]} \\to K(E,w) \\to GW^{[n+1]}$.","pith_inferences":["A consequence the author leaves implicit is that the same quadratic functor should make Grothendieck-Witt groups invariant under derived equivalences, since quasi-isomorphisms are built into the weak equivalences; testing this on a derived equivalence between rings would be a direct check.","The sign conventions in Definition 10.2 are tied to homological indexing, so extending the construction to cohomological indexing or to unbounded complexes would require a separate verification of quadratic left exactness; this is a natural stress test of the framework.","If the promised comparison with infinity-categorical Hermitian K-theory holds, the 1-categorical additivity, fibration, and cofinality theorems would supply the same results for Poincaré infinity-categories without reproving them, and would give an independent check of the infinity-categorical formalism."],"forward_implications":["For any exact form category with strong duality, $GW(E,i,Q) \\simeq GW(\\mathrm{Ch}^b E, \\mathrm{quis}, Q)$, so bounded chain complexes can be used to compute Grothendieck-Witt groups.","The Fibration Theorem produces homotopy cartesian squares when weak equivalences are changed, giving a localisation principle for Hermitian K-theory analogous to algebraic K-theory.","The Cofinality Theorem shows that cofinal duality-closed subcategories induce isomorphisms on positive Grothendieck-Witt groups and a monomorphism on $GW_0$, matching K-theory behaviour.","The Bott sequence $GW^{[n]} \\to K(E,w) \\to GW^{[n+1]}$ deloops Grothendieck-Witt spaces, and the negative homotopy groups of the spectrum are the Witt groups of shifted categories.","Together with the announced comparison to the infinity-categorical theory, the classical 1-categorical Hermitian K-theory would agree with the infinity-categorical version in full generality, going beyond previously known split-exact, 2-invertible, and Zariski-descent cases."],"supporting_citations":[{"why":"Sets up exact form categories with strong duality and defines the Grothendieck-Witt space that this paper extends.","marker":"[Sch21]"},{"why":"Proves the symmetric-forms version of the main theorem, the statement generalised here to arbitrary quadratic forms.","marker":"[Sch10b, Proposition 6]"},{"why":"Supplies the form Q-construction, hyperbolic and forgetful functors, and cofinality results adapted throughout.","marker":"[Sch10a]"},{"why":"Provides the symmetric-form Grothendieck-Witt spectrum and Bott sequence that Section 11 generalises.","marker":"[Sch17]"},{"why":"Gives the S_•-construction and edgewise subdivision underlying the definition of the Grothendieck-Witt space.","marker":"[Wal85]"},{"why":"Supplies Quillen's Q-construction on which the form Q-construction is built.","marker":"[Qui73]"},{"why":"The infinity-categorical Hermitian K-theory whose agreement with the 1-categorical theory the paper's results are designed to establish.","marker":"[CDH+23]"},{"why":"Provides localisation and delooping for exact categories used in the Fibration Theorem.","marker":"[Sch04]"}],"fun_headline_variants":["Bounded complexes preserve Grothendieck-Witt spaces","Quadratic forms on chain complexes keep GW-spaces fixed","Exact categories and their chain complexes share Grothendieck-Witt groups","Chain inclusion is a homotopy equivalence for GW-spaces"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The proof depends on the reduction that passes from $E$ to its semi-idempotent completion without changing Grothendieck-Witt spaces; if that reduction failed, acyclic complexes would not be strictly acyclic and the chain of equivalences would collapse.","fun_headline_variants_meta":{"raw":{"variants":["Bounded complexes preserve Grothendieck-Witt spaces","Quadratic forms on chain complexes keep GW-spaces fixed","Exact categories and their chain complexes share Grothendieck-Witt groups","Chain inclusion is a homotopy equivalence for GW-spaces"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000487,"raw_usage":{"total_tokens":2367,"prompt_tokens":878,"completion_tokens":1489,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":494,"completion_tokens_details":{"reasoning_tokens":1417}},"tokens_in":494,"tokens_out":1489,"duration_ms":12946,"temperature":1.0,"reasoning_tokens":1417,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T21:23:22.818119+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Check the surjectivity step of Lemma 10.4 on an exact category that is not semi-idempotent complete, for instance finitely generated free modules over a ring possessing a non-free stably free module. If some nondegenerate quadratic form on a bounded complex cannot be rewritten as a degree-zero form plus hyperbolics, then $GW_0(E,i,Q) \\to GW_0(\\mathrm{Ch}^b E, \\mathrm{quis}, Q)$ is not surjective and Theorem 10.5 is false for that category.","supporting_citations":[],"review_version":1}