Pith. sign in

REVIEW 4 major objections 5 minor 1 cited by

Parameterized Lagrangian Floer homotopy

T0 review · 4 major / 5 minor · reviewed 2026-08-15 · deepseek-v4-flash

Pith's one-line read In a plumbing of two cotangent bundles, any Hamiltonian deformation of one Lagrangian must meet the other in at least two points of distinct action, even when intersections are degenerate.

desk verdict New parameterized Floer homotopy construction with a plausible and significant application, but the main theorem currently rests on an unproved bridge (Prop. 4.13 / eq. 4.66) that any referee must see filled. read the letter →

arxiv 2506.20122 v2 pith:CKYL4KFB submitted 2025-06-25 math.SG

classification math.SG MSC 53D4053D1255P42
keywords LagrangianFloerhomotopyparameterizedspectraMaslovdataintersectionsplumbingsofcotangentbundlesThomstablesphericalfibrationssphericallyframedbordism
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

Floer theory usually extracts from the intersection problem a cohomology group, which can forget the action-level information carried by individual intersection points. This paper constructs a richer invariant: the spherically framed Lagrangian Floer homotopy type, a spectrum that varies continuously as the Maslov data (the framing choices Floer theory needs) move through their moduli space. The main theorem says that in a plumbing $T^*Q_0 \cup_{N_C} T^*Q_1$ of two cotangent bundles along a connected, orientable, positive-dimensional closed submanifold $C$, any compactly supported Hamiltonian deformation of $L_0$ must meet any such deformation of $L_1$ in at least two points of distinct action, even when the intersections are degenerate. Existing clean-intersection technology already forces two non-degenerate points; the parameterized spectrum sees through degeneracies and pins the surviving intersections to two different action levels.

What carries the argument

The load-bearing object is the parameterized spherically framed Lagrangian Floer homotopy type $F^{\mathrm{sfr}}$, an $\infty$-functor sending each choice of Maslov data $\theta$ — a null-homotopy of the stabilizing map $P(L_0,L_1) \to B^2\mathrm{GL}_1 S$, equivalently a coherent trivialization of the Floer moduli spaces as stable spherical fibrations — to an $R_{\mathrm{sfr}}$-module spectrum assembled from the compactified moduli spaces of Floer trajectories. The ring spectrum $R_{\mathrm{sfr}}$ is the spherically framed analogue of the sphere spectrum: its homotopy groups are bordism classes of manifolds whose tangent bundles are trivial as stable spherical fibrations. The mechanism that carries the argument is the identification of Proposition 4.13: over a plumbing, pulling $F^{\mathrm{sfr}}$ back along $\mathrm{Map}_*(C, B\mathrm{GL}_1 S) \to \mathrm{Mas}_\theta$ recovers the Thom functor $\Phi \mapsto C_\Phi \wedge R_{\mathrm{sfr}}$, with $C_\Phi$ the Thom spectrum of the stable spherical fibration $\Phi$. The non-triviality of this functor on the component of the trivial fibration, proved through negative-degree stable cohomotopy classes supplied by the integral homology of $C$, is what forces intersection points to occupy distinct action levels.

What would settle it

Compute the Floer spectrum for a single non-trivial stable spherical fibration $\Phi$ over $C$, for example the non-trivial line bundle over $S^1$ in the plumbing of two $T^*S^2$'s, and compare it with the corresponding Thom spectrum; if the two spectra differ in homotopy, the bridge labelled (4.66) fails and the proof of the main theorem loses its connection to the spectral non-factorization step. The most direct falsifier of the theorem itself is a compactly supported Hamiltonian pair in a plumbing whose intersection points are finite in number and all share one action level.

Watch

Extended reading notes

Core claim

The central claim, Theorem 1.4, is that the cleanly intersecting Lagrangian pair $(L_0,L_1)$ in the plumbing $T^*Q_0 \cup_{N_C} T^*Q_1$ is rigid: every compactly supported Hamiltonian deformation of $L_0$ intersects every compactly supported Hamiltonian deformation of $L_1$ in at least two points of distinct action, degeneracy allowed. The proof is carried by a new object, the parameterized spherically framed Lagrangian Floer homotopy type $F^{\mathrm{sfr}}\colon \mathrm{Mas}_\theta \to \mathrm{Mod}_{R_{\mathrm{sfr}}}$, an $\infty$-functor from the space of Maslov data to the stable $\infty$-category of modules over the ring spectrum $R_{\mathrm{sfr}}$, whose homotopy groups are the spherically framed bordism groups. In the plumbing setting the natural map $\mathrm{Map}_*(C, B\mathrm{GL}_1 S) \to \mathrm{Mas}_\theta$ pulls $F^{\mathrm{sfr}}$ back to the Thom functor $\Phi \mapsto C_\Phi \wedge R_{\mathrm{sfr}}$ (Proposition 4.13), and Proposition 5.1 shows that on the connected component of the trivial stable fibration this functor cannot factor through a point. If some Hamiltonian deformation achieved a single action level, the whole parameterized functor would factor through a point; that contradiction with Proposition 5.1 is what forces at least two distinct action levels to survive.

Load-bearing premise

The entire proof rests on a single identification, stated with a sketch ('a suitable modification of this...shows') rather than a full proof: that a twisted version of the Floer spectrum, built using a spherical fibration over the intersection locus as extra framing data, is exactly the Thom spectrum associated to that fibration.

Editorial extensions

If this is right

  • For cleanly intersecting Lagrangians, an existing spectral sequence already guarantees two non-degenerate intersection points, and Theorem 1.4 upgrades this to two points of distinct action, so degenerate configurations cannot be compressed onto a single action level.
  • The conclusion is stable under compactly supported Hamiltonian perturbations: no such deformation of the Lagrangian pair in a plumbing can produce a one-point intersection or concentrate all intersections on one action level.
  • Different choices of Maslov data yield genuinely different Floer homotopy types, so non-standard framings detect information — such as the $\Sigma^{-1}\mathbb{RP}^2$ Steenrod square in the $S^1$-plumbing example — that the standard polarization misses, and the parameterized spectrum makes this dependence explicit.
  • Orientability of $C$ enters only to secure a $\mathbb{Z}$-summand in $H^*(C;\mathbb{Z})$; the same contradiction should run over Morava $K$-theory at $p=2$, so the two-level conclusion is expected to survive for non-orientable $C$, as the paper itself suggests.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The mechanism appears transferable: any Lagrangian pair whose path space maps to the free $E_1$-group on the clean intersection locus, so that $\mathrm{Map}_*(C, B\mathrm{GL}_1 S)$ embeds into the space of Maslov data, should inherit a comparable two-level lower bound; plumbings are the instance worked out here, not the only possible one.
  • A concrete test case is the plumbing along $C=S^1$ inside two $T^*S^2$'s, where the parameter space $\mathrm{Map}_*(S^1, B\mathrm{GL}_1 S)$ is explicit enough that the non-constant family of Floer spectra could be written down, making the distinct-action phenomenon visible by direct computation.
  • Nothing in the argument singles out the number two: feeding the same non-factorization mechanism with additional non-trivial stable cohomotopy classes suggests that a richer intersection locus $C$ could force more than two action levels, although the paper itself claims only two.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 5 minor

Summary. The paper constructs a parameterized version of the spherically framed Lagrangian Floer homotopy type for exact Lagrangians in Liouville manifolds satisfying a Maslov-data condition. The main object is an ∞-functor F^{sfr}: Mas_θ → Mod_{R_sfr} (eq. 1.9), and the main application is Theorem 1.4: for a plumbing of two cotangent bundles along a closed connected orientable positive-dimensional manifold C, any compactly supported Hamiltonian deformation of the two Lagrangians yields at least two intersection points with distinct action. The proof proceeds by showing that after pulling back F^{sfr} to Map_*(C, BGL_1 S) it is the Thom functor Φ ↦ C_Φ ∧ R_sfr (Prop. 4.13), proving that this pulled-back functor cannot factor through a point (Prop. 5.1), and then arguing that a single-action intersection would force such a factorization.

Significance. If the construction and identification are correct, this is a significant advance: it gives lower bounds for degenerate Lagrangian intersections beyond the reach of ordinary Floer cohomology, and it demonstrates that the choice of Maslov data carries nontrivial homotopical information. The stable-homotopy core of the non-factorization argument (Lemma 5.2 and Corollary 5.3) is clean and appears correct, and the idea of using the spherically framed bordism ring spectrum R_sfr to convert a non-trivial top-right corner into infinitely many non-zero maps is appealing. The paper is also honest about its conjectural parts. However, the main theorem is not fully proven as written: the identification (4.66) is the bridge from the stable-homotopy computation to Lagrangian intersection theory, and that bridge is currently not constructed.

major comments (4)
  1. [§4.2, Prop. 4.13, eq. (4.66)] The homotopy equivalence F_{H,J,θ_Φ} ≃ C_Φ ∧ R_sfr is the central bridge of the paper, but its proof is delegated to 'a suitable modification of this...shows' and is not carried out. In particular, eq. (4.69) adds the stable spherical fibration Φ|_{y(0)} to the real vector space R^{μ(y)} and Φ|_{x(0)} to R^{μ(x)}; this operation is not defined for real virtual bundles, since an arbitrary stable spherical fibration need not lie in the image of the J-homomorphism BO → BGL_1S. The framework of Definition 2.6 only allows a twist of the spherical trivialization of the virtual bundle I(x,y), not a shift of the real vector spaces V_x and V_y. Until a real virtual bundle realizing the Φ-twist is constructed, or the flow-category formalism is generalized, the identification of the pullback with the Thom functor is not established. Since Proposition 5.1 and the contradiction in §5.2 depend on this identification, this is a load-bearing gap, though it appears to be a gap rather than a demonstrated falsehood.
  2. [§5.2, factorization claim (5.41)] The assertion that a single action level makes the ∞-functor (5.41) factor through a constant functor is not proven. The supporting parenthetical about components of P(L0,L1) is vacuous: for the plumbing setup, P(L0,L1) is connected, so there are no 'other components' whose Floer homotopy types are trivial. Moreover, a factorization of an ∞-functor through a point requires compatible coherence data for all higher simplices, not just null-homotopies of the 1-dimensional maps to BGL_1S as discussed after (5.36)–(5.39). This step is essential to obtain the contradiction, so it needs a complete argument.
  3. [App. A and App. B, Cor. A.3 and Thm B.8, eq. (B.15)] The nonvanishing input for Proposition 5.1 is not correctly established because of a grading confusion. With the standard convention \tilde{S}^j(X) = [X, S^j] used in the paper (cf. (B.4)), the groups for X = S^n and j ≤ 0 vanish, so Corollary A.3's claim of a non-trivial reduced stable cohomotopy class in non-positive degree is false as stated. Similarly, eq. (B.15) appeals to π_j S_(p), which is zero for j < 0, in order to prove non-vanishing of the map (B.14) for infinitely many negative j. The proof of Theorem B.8 therefore does not establish the needed non-vanishing; a correct formulation using Spanier-Whitehead duality or the appropriate positive cohomological degrees is required.
  4. [§5.2, eq. (5.34)] The proof of Theorem 1.4 assumes without justification that the indexing set A in (5.34) is finite. This is not automatic for degenerate intersections: compactly supported Hamiltonian deformations can have positive-dimensional intersection loci (e.g., clean intersections), and the hypothesis that all intersection points have the same action does not force finiteness. The subsequent wedge decomposition in (5.42) and the perturbation argument require a finite generating set; the authors should either justify finiteness or replace the argument by a Morsification of the local generating functions that produces finitely many nondegenerate intersection points while controlling the action levels.
minor comments (5)
  1. [Lemma B.2 proof] In the proof of Lemma B.2, 'fibrationa' should be 'fibration'.
  2. [Abstract and §1] The notation 'Masθ' in the abstract and 'Mas_θ' elsewhere should be unified.
  3. [Prop. 4.13 proof] The equality Ψ(Φ)(γ_c) = Φ|_c relies on a choice of basepoint in C and on the free E_1-algebra property of ΩΣC; a few clarifying words about the basepoint conventions would help.
  4. [Remark 1.6] The proposed Morava K-theory variant would require a definition of Maslov data with BGL_1K(n); the sentence is difficult to parse and would benefit from being spelled out.
  5. [References] Reference [Bla] is listed as an unpublished note; if it is cited for the Steenrod-square lower bounds in §1.1.1, please indicate its availability or state that it is a personal communication.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity found; the main theorem is proved by a contradiction whose nontrivial input is an independent stable-homotopy calculation.

full rationale

The derivation is not circular. The proof of Theorem 1.4 runs a genuine contradiction argument: assuming a Hamiltonian deformation with all intersection points at a single action level, Section 5.2 derives that the infinity-functor (5.41) factors through a constant functor, while Proposition 5.1 shows by a concrete stable-homotopy computation (Lemma 5.2, Corollary 5.3, and Theorem B.8) that the same functor cannot factor through a point. Neither the Thom-spectrum computation nor the bordism split-injection in Appendix B assumes the desired intersection lower bound. The construction of the parameterized Lagrangian Floer homotopy type and its identification with the Thom functor in Proposition 4.13 rely on prior work of the authors, notably [Bla24] for the Morse-theoretic reduction in plumbings and [Bon25] for the Bott isomorphism, but those are technical inputs about the geometry and index theory, not restatements of Theorem 1.4; self-citation is therefore not load-bearing in the circularity sense. The only notable weakness is a correctness gap rather than a circular step: the bridge equivalence (4.66) is asserted with 'a suitable modification of this ... shows', and equation (4.69) appears to add a stable spherical fibration to real vector spaces in a way that is not generally well-defined without first realizing the Phi-twist as a real virtual bundle. If (4.66) fails, Proposition 5.1 would not transfer to the Lagrangian Floer homotopy types, but that is an unproven step in the derivation, not a reduction of the theorem to its own inputs. No fitted parameter is renamed as a prediction, and no uniqueness theorem from the authors' earlier work is invoked to force the chosen framing.

Assumptions & free parameters 0 free parameters · 6 assumptions · 1 invented entities

The paper pulls in a large amount of background from AB24, Lar21, Bla24, and Bon25. The entries flagged as axioms are the load-bearing inputs not fully paid for in this text: Assumption 1 (the geometric setting), the assertion that π_*R_sfr is spherically framed bordism (Remark 2.8), the compatibility of Bonciocat's Bott isomorphism with CR index theory (Props 3.3 and 4.9), Theorem A.1's p-local splitting, the unproven 'suitable modification' identification in Prop 4.13, and the §5.2 factorization claim. No free parameters are fitted to data; this is a proof paper. The one invented entity is R_sfr itself, whose identification with a classical spectrum is left conjectural (Conjectures B.4 and B.5).

assumptions (6)
  • domain assumption Assumption 1: M is a Liouville manifold, L_i is exact and closed or cylindrical at infinity, and (L_0,L_1) admits Maslov data θ (a null-homotopy of the map (1.8)).
    This is the standing setup for the whole construction (Section 1.1.2, Assumption 1). The plumbing examples in Theorem 1.4 are asserted to satisfy it via the standard stable R-polarization and a suitable basepoint θ (proof of Prop 4.13).
  • standard math The Abouzaid-Blumberg flow-category framework, including R_fr ≃ S and the assertion that π_*R_sfr is the spherically framed bordism ring (Remark 2.8).
    Invoked throughout (Section 2); the π_*R_sfr identification is used in Propositions B.6 and B.8, which feed Proposition 5.1.
  • ad hoc to paper The Bott isomorphism of [Bon25] is compatible with the index theory of Cauchy-Riemann operators with totally real boundary conditions, including the Eckmann-Hilton compatibility (intertwining addition of index bundles with loop concatenation).
    The final sentence of the proof of Proposition 3.3 and the proof of Proposition 4.9 rest on this; it is cited to arXiv:2506.17431 by the second author, submitted two days before this paper, and not demonstrated here.
  • standard math Theorem A.1: for a finite spectrum X and sufficiently large prime p, X_(p) is a wedge of shifted Moore spectra.
    Proven in Appendix A and used in Corollary A.3 and Theorem B.8 to produce the negative-degree stable cohomotopy classes.
  • ad hoc to paper The identification FH,J,θΦ ≃ C_Φ ∧ R_sfr (Prop 4.13, eq. 4.66), including the unshown 'suitable modification' for the twist by Φ.
    This asserted equivalence is the bridge from the Floer construction to the Thom functor; Theorem 1.4's contradiction uses it to import the non-factorization of Prop 5.1. The surrounding text (eqs. 4.56-4.69) outlines the mechanism but does not complete the proof.
  • ad hoc to paper The §5.2 claim: if all intersection points share one action level, the parameterized ∞-functor (5.41) factors through a constant functor, including all higher coherence data.
    The 'upshot' paragraph asserts this. The supporting parenthetical about components of P(L_0,L_1) other than the c_0 component is vacuous because P(L_0,L_1) is connected; the ∞-categorical coherence is not shown.
invented entities (1)
  • R_sfr, the spherically framed bordism ring spectrum (Flow^{sfr}(1,1))
    purpose: Coefficient ring for the parameterized Floer homotopy type; the module category Mod_{R_sfr} is the target of the ∞-functor (1.9), and R_sfr's negative-degree classes drive Proposition 5.1 via Theorem B.8.
    Defined inside the flow-category framework, but its identification with a classical spectrum (e.g., Σ^∞ hofib BJ) is left as Conjecture B.5, and the homology-theory identification is Conjecture B.4. The main theorem depends on properties of R_sfr asserted at sketch level, so there is no independent falsifiable handle in this paper.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Parameterized Lagrangian Floer homotopy." pith.science (2026). https://pith.science/paper/CKYL4KFB

@misc{pith2026250620122,
  author       = {Pith},
  title        = {Pith review of: Parameterized Lagrangian Floer homotopy},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/CKYL4KFB}},
  note         = {Machine review of arXiv:2506.20122}
}
read the original abstract

We construct the Lagrangian Floer homotopy type, in the exact setting, as a spectrum parameterized over the moduli space of Maslov data. Our primary motivation for this construction is to provide stronger lower bounds for (possibly degenerate) Lagrangian intersections in plumbings of cotangent bundles.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Ample divisor complements, Floer spectra, and relative Gromov-Witten theory

    math.SG 2026-01 conditional novelty 7.0 of 10

    The associated graded of the Floer homotopy type of an ample smooth divisor complement is computed, with the splitting obstruction encoded in a stable homotopy class from genus-0 relative Gromov–Witten moduli.

Reference graph

Works this paper leans on

22 extracted references · 18 canonical work pages · cited by 1 Pith paper

  1. [1]

    Abouzaid and A

    M. Abouzaid and A. Blumberg. Foundation of Floer homotopy theory I: flow categories . arXiv:2404.03193, 2024

  2. [2]

    Asplund, Y

    J. Asplund, Y. Deshmukh, and A. Pieloch. Spectral equivalence of nearby Lagrangians . arXiv:2411.08841, 2024

  3. [3]

    \'Alvarez-Gavela, T

    D. \'Alvarez-Gavela, T. Large, and A. Ward. On arborealization, Maslov data, and lack thereof . arXiv:2503.09783, 2025

  4. [4]

    K. Blakey. Dual Steenrod algebras and the spectral quantum cap product . unpublished note available at kenneth-blakey.com

  5. [5]

    K. Blakey. Floer homotopy theory and degenerate Lagrangian intersections . arXiv:2410.11478, 2024

  6. [6]

    Revisiting the Cohen-Jones-Segal construction in Morse-Bott theory

    C.M. Bonciocat. Revisiting the Cohen-Jones-Segal construction in Morse-Bott theory . arXiv:2409.11278, 2024

  7. [7]

    Floer homotopy theory for monotone Lagrangians

    C.M. Bonciocat. Floer homotopy theory for monotone Lagrangians . arXiv:2506.17431, 2025

  8. [8]

    Cohen, J

    R. Cohen, J. Jones, and G. Segal. Floer's infinite dimensional Morse theory and homotopy theory . The Floer Memorial Volume, Progress in Mathematics , 133:297--325, 1995

Show all 22 references
  1. [9]

    C \^o t \'e and Y

    L. C \^o t \'e and Y. B. Kartal. Equivariant Floer homotopy via Morse-Bott theory. arXiv:2309.15089, 2023

  2. [10]

    A. Floer. Morse theory for Lagrangian intersections . Journal of Differential Geometry , 38(3):513--547, 1988

  3. [11]

    A. Floer. The unregularized gradient flow of the symplectic action. Communications on Pure and Applied Mathematics , 41(6):775--813, 1988

  4. [12]

    A. Floer. Cuplength estimates on Lagrangian intersections . Communications on Pure and Applied Mathematics , 42(4):335--356, 1989

  5. [13]

    H. Hofer. Lusternik-Schnirelman-theory for Lagrangian intersections . Annales de l'Institut Henri Poincar\'e. Analyse non lin\'eaire , 5(5):465--499, 1988

  6. [14]

    Hirschi and N

    A. Hirschi and N. Porcelli. Lagrangian intersections and cuplength in generalised cohomology theories. To appear in Mathematical Research Letters, 2022

  7. [15]

    T. Large. Spectral Fukaya categories for Liouville manifolds. MIT Ph.D. Thesis, 2021

  8. [16]

    G. Laures. On cobordism of manifolds with corners. Transactions of the American Mathematical Society , 352(12):5667--5688, 2000

  9. [17]

    Laudenbach and J.-C

    F. Laudenbach and J.-C. Sikorav. Persistance d'intersection avec la section nulle au cours d'une isotopie hamiltonienne dans un fibré cotangent . Inventiones mathematicae , 82:349--357, 1985

  10. [18]

    Po \'z niak

    M. Po \'z niak. Floer homology, Novikov rings, and clean intersection . University of Warwick Ph.D. Thesis, 1994

  11. [19]

    Porcelli and I

    N. Porcelli and I. Smith. Bordism of flow modules and exact Lagrangians . arXiv:2401.11766, 2024

  12. [20]

    Porcelli and I

    N. Porcelli and I. Smith. Spectral Floer theory and tangential structures . arXiv:2411.03257, 2024

  13. [21]

    P. Seidel. Graded Lagrangian submanifolds . Bulletin de la Soci\'et\'e Math\'ematique de France , 128(1):103--149, 2000

  14. [22]

    H.L. Tanaka. Functors (between -categories) that aren't strictly unital . Journal of Homotopy and Related Structures , 13:237--286, 2018

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.