{"id":"f860633b-cec3-4310-824a-54a7317f2df8","arxiv_id":"2501.11892","paper_version":1,"verdict":"ACCEPT","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"A new gluing theorem for parameterized Seiberg-Witten invariants gives infinite rank Z^∞ summands in higher homotopy and homology of diffeomorphism groups of 4-manifolds that are topologically trivial.","lead":"This paper builds infinite families of smooth symmetries of certain four-dimensional spaces that have no topological counterpart, detecting them with a new gauge-theoretic counting tool. The result shows that in dimension four, families of diffeomorphisms can be far richer than families of homeomorphisms, with consequences for spaces of embeddings and positive scalar curvature metrics.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The proof of Theorem 1.1 hinges on Theorem 5.4; the least secure step is Proposition 9.5's reduction to locally metric-independent data, whose higher-parameter proof is only sketched and asserts an exact preservation of the moduli space.","rationale":"The paper is a substantial and internally consistent research preprint, and the reader's identification of Theorem 5.4 as the main analytical ingredient is correct. My stress-test narrows this to two specific places inside Theorem 5.4: Proposition 9.5's reduction to locally metric-independent data, and Lemma 9.8's parameterized linear gluing. Both are presented as adaptations of cited results ([MRS11], [Nic02]) rather than with fully detailed proofs, yet both are essential for the gluing theorem to apply to the specific families arising in the suspension construction. If these analytic reductions fail, the computation of the family Seiberg-Witten invariants for the commutator families is unsupported, and Theorem 1.1 loses its proof. This does not mean the theorem is false; the arguments are plausible and the rest of the paper is well structured. But because the central claim depends so directly on this unverified analytic step, acceptance should be conditional on a detailed independent check of Proposition 9.5 and Lemma 9.8. I found no other comparable obstruction: the base-case construction, the mod 2 to integer passage, and the applications to embeddings and PSC metrics all appear to follow once Theorem 5.4 is available.","tokens_in":63252,"tokens_out":36556,"duration_ms":383806,"concrete_test":"Independently verify Proposition 9.5 in the k-parameter setting: starting from good data on a k-dimensional parameter space with at least two exceptional points, perform the ball-wise metric modification described in Section 9.2.1 and check, using the [KM08] transversality framework, that the mod-2 count of the zero-dimensional parameterized moduli space is unchanged and that no solutions are created or destroyed. Separately, write out the proof of Lemma 9.8 with the finite-rank term n(θ) included, checking the claimed uniform estimates on the neck where n vanishes. If either check fails, Theorem 5.4 is not established in the form needed for Corollary 5.6 and Suspension Theorem 5.7.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The Suspension Theorem 5.7 reduces the computation of SW invariants for the commutator families to Corollary 5.6, which is proved using the Parameterized Irreducible-Reducible Gluing Theorem 5.4. If Theorem 5.4 fails, the recursive calculation in Theorem 5.8, and with it the Z^∞ summands of Theorem 1.1, collapses. The theorem is therefore genuinely load-bearing. Within its proof, the step that lets the parameterized problem be treated like the unparameterized one is Proposition 9.5: for a zero-dimensional parameterized moduli space with exceptional points in top-dimensional simplices, it claims that a small deformation makes the data locally metric-independent while keeping the moduli space exactly the same. The proof is a sketch: it says a version is proved in [MRS11, Appendix A] for 1-parameter families and 'adapts readily' to k-dimensional parameter spaces; it works locally around one exceptional point and asserts no new solutions appear. It does not spell out how several exceptional points are handled simultaneously, why the modified metric cannot create or destroy solutions on the perturbed region, or why the resulting data remains good globally. A second condensed analytical input is Lemma 9.8, where parameterized linear gluing with a finite-rank perturbation n(θ) is asserted as a modification of [Nic02]; the estimates are summarized rather than written out. No internal contradiction was found, but these unverified details are the soft spot in the central argument.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper constructs spherical families of diffeomorphisms of certain 4-manifolds and shows that they give Z^∞ summands in π_j(Diff_0(Z_p)) and H_j(Diff_0(Z_p)) which lie in the kernel of the natural maps to the corresponding groups for Homeo_0(Z_p); analogous statements are obtained for H_{j+1}(B TDiff(Z_p)). The construction is inductive, starting from the second author's 1998 non-isotopic diffeomorphisms, and the non-triviality is detected by family Seiberg-Witten invariants. The central analytic input is a new parameterized irreducible-reducible gluing theorem (Theorem 5.4), from which a suspension theorem (Theorem 5.7) and concrete invariant computations (Theorem 5.8) are derived. Applications include infinite generation in homotopy and homology of embedding spaces of S^2, S^3, and S^1×S^2, and of spaces of positive scalar curvature metrics.","tokens_in":63549,"tokens_out":6237,"duration_ms":64242,"significance":"If the analytic core is valid, the paper is a substantial advance: it establishes infinite-rank summands in higher homotopy and homology groups of diffeomorphism groups of 4-manifolds that vanish topologically, in contrast with known finiteness results in higher dimensions, and it gives new applications to embedding spaces and PSC metrics. The paper is commendably explicit about its limitations, e.g., Remark 5.5 notes that the gluing theorem is proved only for zero-dimensional parameterized moduli spaces and without a full orientation analysis. The recursive construction is concrete, and the paper correctly explains the passage from mod 2 to integer invariants via Lemma 2.9 and Lemma 2.21. The main risk is the depth of the analytic gluing argument, which I could not independently verify in full detail.","major_comments":[{"comment":"The proof of Proposition 9.5, which deforms good data to locally metric independent data, is the principal unverified step in the paper's analytic core. The text states that a version is proved in [MRS11, Appendix A] for 1-parameter families and that the argument 'adapts readily' to k-dimensional parameter spaces, and the proof given works locally around one exceptional point in a top-dimensional simplex. It is not explained how several exceptional points in different simplices are treated simultaneously, why the modified metric cannot create or destroy solutions on the perturbed region, or why the resulting globally defined data remains irreducible-reducible good. Since Proposition 9.5 is used in the proof of Theorem 5.4 and then in Corollary 5.6 and the Suspension Theorem 5.7, this is load-bearing. Please provide a complete proof or a reference containing exactly this k-parameter, multi-point statement, rather than an assertion of adaptation.","section":"9.2.1, Proposition 9.5"},{"comment":"The parameterized linear gluing result Lemma 9.8 is asserted as a modification of [Nic02] with the proof summarized in a single paragraph. In particular, the treatment of the finite-rank perturbation n(θ) inside the operator rD, the claimed bounds used in the bootstrapping argument, and the exactness of the displayed sequence with the new parameter summand are not written out. Because this lemma is the linearization step underlying Theorem 5.4, a failure or gap here would invalidate the gluing theorem and hence the main computation. The proof should be expanded to the same level of detail as the rest of Section 9, or a precise reference covering this parameterized perturbation should be supplied.","section":"9.3.2, Lemma 9.8"}],"minor_comments":[{"comment":"The notation 'Z8' is used for what is clearly meant to be an infinite direct sum of copies of Z; please use \\mathbb{Z}^\\infty or \\bigoplus_{\\mathbb{N}}\\mathbb{Z} to avoid confusion with the cyclic group of order 8, which appears nowhere else in the paper.","section":"Abstract and throughout"},{"comment":"The abstract contains the typo 'we we obtain'; please correct it.","section":"Abstract"},{"comment":"In the paragraph following Definition 2.8, 'For n = 0' should be 'For k = 0', since the parameter sphere is S^k throughout the section.","section":"Section 2.4.1"},{"comment":"The statement of Proposition 9.5 does not explicitly record the hypothesis b_2^+(X) > dim(Ξ), although the proof and the surrounding discussion use this to ensure that the family data can be chosen good; please add this condition to the statement.","section":"Proposition 9.5"},{"comment":"The proof of Corollary 5.6 is very terse: it says the parameterized moduli space is the product of the moduli space on Z and the one-point moduli space for S^2×S^2, but it does not explicitly invoke the identification of the S^1-fibered product with the ordinary product when one factor is a point; a sentence making this identification would improve readability.","section":"Section 9.4, proof of Corollary 5.6"}],"recommendation":"major_revision","confidential_remarks":"The paper is technically impressive and likely correct, but the two load-bearing analytic steps identified in the major comments (Proposition 9.5 and Lemma 9.8) are precisely the places where the proof is sketched rather than written out. Given that the entire computation of the Seiberg-Witten invariants collapses if either of these steps fails, I would recommend that the editor seek a careful review of Section 9 by a specialist in analytic gluing theory for Seiberg-Witten moduli spaces. The reliance on the companion paper [AR23], described as 'in preparation', is not heavy in the current text, but it would be helpful if the authors could clarify which results of the main theorem, if any, depend on [AR23] beyond the motivational remarks."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Here's my take. The headline result is real: for each p>0 they build 4-manifolds Z_p with Z^∞ summands in π_j and H_j of Diff_0, lying in the kernel of the comparison to Homeo_0, for j ≡ p mod 2 up to p. This contrasts sharply with high-dimensional finiteness and answers a question Lin highlighted. The applications to embedding spaces and PSC metrics are natural and extend earlier work. The paper is honest: it explicitly says the mod 2 to integer step is delicate, and it doesn't pretend Theorem 5.4 is routine.\n\nWhat's new: the parameterized irreducible-reducible gluing theorem (Theorem 5.4) allowing different parameter spaces on the two summands is a genuine extension of Baraglia–Konno. The spin examples and the recursive suspension scheme are also new.\n\nWhat's good: the paper is careful. Section 9 actually proves the gluing theorem with analytic detail. The construction is self-contained except for the companion [AR23], which is used for the conceptual frame but not for the main analytic inputs. The mod 2 counting is handled cleanly via Lemma 2.9, and the vanishing result (Lemma 2.21) is used to get integer summands from Z_2 non-vanishing. The presentation is transparent about what is proved and what is imported.\n\nSoft spots: the load-bearing step is Theorem 5.4, and the least written-out part is Proposition 9.5, which asserts that zero-dimensional parameterized moduli spaces can be deformed to locally metric-independent data while preserving the moduli space exactly. The proof is a sketch: it cites [MRS11] for the 1-parameter case and says it 'adapts readily' to k parameters. It doesn't spell out handling several exceptional points simultaneously or why the perturbation can't create new solutions. Lemma 9.8 (parameterized linear gluing with finite-rank perturbation) is also summarized rather than proved. These are not contradictions—they're places where a referee needs to do real work. If Theorem 5.4 fails, the recursive computation collapses, so this is genuinely load-bearing. But I see no evidence it fails; the sketch points to standard methods and the structure is plausible.\n\nAnother caveat: the paper relies on the unpublished [AR23] for the recursive scheme, but the current paper's arguments are self-contained enough that I don't see this as a fatal issue.\n\nBottom line: this is a paper for specialists in 4-manifold gauge theory and diffeomorphism groups. It deserves a serious referee. The editor should send it to review, with the request that the referee scrutinize Section 9, especially Proposition 9.5. I would not desk-reject it; it's important and likely correct.","headline":"Ambitious and likely correct paper giving Z^∞ summands in higher homotopy/homology of diffeomorphism groups that die topologically; the analytical gluing is the main thing to check.","tokens_in":64072,"tokens_out":3344,"would_cite":true,"duration_ms":34783,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["57K41","57R52","53C21"],"pacs":[],"model":"deepseek-v4-flash","headline":"Seiberg-Witten theory builds infinite-rank families of diffeomorphisms invisible to homeomorphisms","keywords":["4-manifolds","diffeomorphism groups","family Seiberg-Witten invariants","homeomorphism groups","Torelli group","positive scalar curvature metrics","embedding spaces","gluing theorem"],"falsifier":"Compute the mod-2 family Seiberg-Witten invariant of one claimed generator α^p[q] in the $\\mathrm{Spin}^c$ structure $K+2\\Sigma\\tilde T+2\\ell\\tilde T_1$ with $|\\ell|\\leq q$; a value of 0 where the recursive formula predicts 1 would disprove the gluing theorem and with it the construction. Equivalently, exhibit irreducible-reducible good data on a connected sum for which the parameterized moduli space is not the $S^1$-fibered product.","tokens_in":63055,"feed_emoji":"🧩","tokens_out":10320,"duration_ms":100011,"temperature":0.7,"pith_summary":"This paper establishes that the smooth and topological categories in dimension four differ sharply at the level of families: for every p>0 there are 4-manifolds whose diffeomorphism groups contain infinite-rank free abelian summands in homotopy and homology groups in a range of degrees, all lying in the kernel of the natural map to the homeomorphism group. The construction is recursive, starting from non-isotopic but pseudoisotopic diffeomorphisms on elliptic surfaces and producing spherical families of diffeomorphisms of arbitrarily high dimension. Non-triviality is detected by family Seiberg-Witten invariants, computed through a new gluing theorem adapted to the inductive construction. The same families yield infinite-rank summands in the homotopy and homology of spaces of embeddings of spheres and 3-manifolds, and in spaces of positive scalar curvature metrics on standard PSC 4-manifolds.","feed_headline":"Infinite-rank exotic diffeomorphism families found in 4-manifolds","feed_subtitle":"Smooth spherical families vanish topologically but show up as infinite-rank summands in homotopy and homology.","key_machinery":"The load-bearing object is the parameterized family Seiberg-Witten invariant, in its homotopy form SWπ_k and homology form SWH_k, together with the new Parameterized Irreducible-Reducible Gluing Theorem (Theorem 5.4). That theorem computes the mod-2 family invariant of a connected sum N_1 # N_2 as an $S^{1}$-fibered product of the family moduli spaces on the two summands, one carrying only isolated irreducible solutions and the other only isolated reducible solutions; a Suspension Theorem (Theorem 5.7) then turns a k-dimensional family on Z into a (k+1)-dimensional family on Z # ($S^{2}$×$S^{2}$) while preserving the invariant. The commutator construction with the reflection diffeomorphism on $S^{2}$×$S^{2}$ raises the sphere dimension, and the finite-set-of-basic-classes vanishing result converts mod-2 detection into infinite rank.","core_discovery":"The central discovery is Theorem 1.1: for any p>0 there are 4-manifolds Z_p such that for all 0<j≤p with j≡p mod 2, the groups π_j(Diff_0(Z_p)) and H_j(Diff_0(Z_p)) contain Z^∞ summands lying in the kernel of the comparison map to the corresponding homeomorphism groups, and likewise for H_{j+1}(B TDiff(Z_p)). The proof builds spherical families by a commutator construction: a k-dimensional family on one manifold is converted into a (k+1)-dimensional family on a stabilized manifold, using reflection diffeomorphisms on $S^{2}$×$S^{2}$ summands and the fact that the seed diffeomorphisms become isotopic after one stabilization. The non-vanishing is established by family Seiberg-Witten invariants, with a mod-2 gluing computation promoted to integer-valued invariants through the composition law, and infinite rank follows because only finitely many basic classes can contribute to each family.","pith_inferences":["As an extension, tracking orientations through Theorem 5.4 would likely yield an integer-valued gluing formula directly, letting the Z^∞ summands be seen without the mod-2-to-integer composition step that the paper currently needs.","As an extension, because the constructed families are supported away from the distinguished nucleus and are stably trivial, the same recursive construction should transplant to any manifold obtained by fiber-summing along that nucleus, potentially making the phenomenon generic among sufficiently stabilized 4-manifolds.","As an extension, the paper leaves open what happens in the quotient R_+(Z)/Diff(Z) of the PSC-metric space; evaluating the same family invariants on the homotopy quotient would test whether the infinite-rank classes survive after forgetting the diffeomorphism parameter.","As an extrapolation, the recursive suspension construction suggests a stable-range phenomenon: the same manifold carries summands in all degrees j≡p mod 2 up to p, and one might expect infinite generation to persist or concentrate in a stable limit as p grows."],"forward_implications":["For each parity class of degrees up to p, the diffeomorphism group of Z_p has a free abelian summand of countably infinite rank that becomes trivial in the homeomorphism group, so the smooth and topological classifications of 4-manifold families diverge in every such degree.","An infinite-rank subgroup of the constructed families becomes smoothly trivial after a single connected sum with S^2×S^2, and each element is topologically isotopic, indeed pseudoisotopic, to the identity.","The sphere bundles built by clutching along these families are smoothly non-trivial but topologically trivial, and their total spaces are diffeomorphic to products.","The same invariants give infinite-rank summands in the homotopy and homology of embedding spaces of S^2, S^3, and S^1×S^2 in stabilized manifolds, and the generators that become trivial after one external stabilization are concordant.","Spaces of positive scalar curvature metrics on standard spin and non-spin PSC 4-manifolds acquire infinite-rank summands in their homotopy and homology groups, including spin examples not covered by earlier methods."],"supporting_citations":[{"why":"Supplies the seed non-isotopic but pseudoisotopic diffeomorphisms and the π0 Seiberg-Witten obstruction that starts the recursion.","marker":"[Rub98]"},{"why":"Provides the 1-parameter Seiberg-Witten invariants and wall-crossing formula that the family invariants generalize, plus the positive scalar curvature obstruction used in Section 8.","marker":"[Rub01]"},{"why":"Gives the prior families gluing formula and the computation of the reflection diffeomorphism invariants on S^2×S^2 that the new gluing theorem extends.","marker":"[BK20]"},{"why":"Provides the cylindrical-end analysis, the blow-up formula, and the neck-stretching framework on which the gluing theorem is built.","marker":"[Nic00]"},{"why":"Establishes stable isotopy of surfaces after one stabilization, used to show the seed diffeomorphisms and later families become trivial after one S^2×S^2 sum.","marker":"[AKMR15]"},{"why":"Provides the log-transformed elliptic surfaces and the diffeomorphisms after one stabilization that distinguish the smooth structures.","marker":"[Gom91]"},{"why":"Defines the family Seiberg-Witten characteristic classes on classifying spaces used for the H_{j+1}(B TDiff) statements.","marker":"[Kon21]"},{"why":"Supplies the spin symplectic 4-manifolds used as building blocks in the recursive construction of the manifolds Z_p.","marker":"[Par02]"},{"why":"Provides the log-transform formula used to compute Seiberg-Witten invariants of the seeded manifolds.","marker":"[FS97]"},{"why":"Gives the non-vanishing of the Seiberg-Witten invariant on the canonical class of symplectic manifolds, anchoring the recursive computation.","marker":"[Tau94b]"}],"fun_headline_variants":["Infinite-rank diffeo families from Seiberg-Witten invariants","Diffeo groups get Z^∞ summands; homeo groups don't","Seiberg-Witten yields infinite-rank families in 4-manifold diffeo groups","Infinite-rank homotopy summands in diffeo groups of 4-manifolds","Exotic diffeo families: infinite rank, vanish topologically"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"Everything rests on the analytic gluing theorem: if the family moduli space of a connected sum is not the glued product of the two pieces' moduli spaces for irreducible-reducible good data, the recursive computation of the Seiberg-Witten invariants collapses.","fun_headline_variants_meta":{"raw":{"variants":["Infinite-rank diffeo families from Seiberg-Witten invariants","Diffeo groups get Z^∞ summands; homeo groups don't","Seiberg-Witten yields infinite-rank families in 4-manifold diffeo groups","Infinite-rank homotopy summands in diffeo groups of 4-manifolds","Exotic diffeo families: infinite rank, vanish topologically"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001151,"raw_usage":{"total_tokens":4788,"prompt_tokens":981,"completion_tokens":3807,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":597,"completion_tokens_details":{"reasoning_tokens":3698}},"tokens_in":597,"tokens_out":3807,"duration_ms":26029,"temperature":1.0,"reasoning_tokens":3698,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T17:44:41.970397+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the mod-2 family Seiberg-Witten invariant of one claimed generator α^p[q] in the $\\mathrm{Spin}^c$ structure $K+2\\Sigma\\tilde T+2\\ell\\tilde T_1$ with $|\\ell|\\leq q$; a value of 0 where the recursive formula predicts 1 would disprove the gluing theorem and with it the construction. Equivalently, exhibit irreducible-reducible good data on a connected sum for which the parameterized moduli space is not the $S^1$-fibered product.","supporting_citations":[],"review_version":1}