Pith. sign in

REVIEW 4 minor 14 references

Real sutured Heegaard Floer homology

T0 review · 0 major / 4 minor · reviewed 2026-07-14 · grok-4.5

Pith's one-line read Real sutured Heegaard Floer homology is well-defined, combinatorially computable, and obeys surface and arc decomposition formulae that mirror and refine the classical theory.

desk verdict Solid, carefully written extension of real HF to the sutured setting that delivers combinatorial computability and new structural theorems; the soft spots are real but not load-bearing. read the letter →

arxiv 2607.11082 v1 pith:T7VG6GN6 submitted 2026-07-13 math.GT

classification math.GT MSC 57K3057K3157R58
keywords realHeegaardFloerhomologysuturedmanifoldsSpin^cstructuresnicediagramssurfacedecompositionsarccombinatorial
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

The paper builds a Floer homology for three-manifolds that come with an orientation-preserving involution of fixed-point set of codimension two, and for their sutured versions. The resulting real sutured Floer homology is independent of the choice of real Heegaard diagram and of a generic symmetric almost-complex structure, splits over real Spin^c structures, and can be read off combinatorially once a nice real diagram is produced. It satisfies an isomorphism under taut real surface decompositions that is formally identical to the classical surface-decomposition formula, and it satisfies a new formula under real arc decompositions that has no classical counterpart. At the same time the theory fails to obey a Künneth formula for connected sums and does not detect real genus or real fibrehood in the way ordinary sutured Floer homology detects ordinary genus and fibrehood. The construction therefore supplies a computable invariant that captures both the classical topology of the underlying manifold and the new constraints imposed by the involution.

What carries the argument

Nice real Heegaard diagrams: real diagrams in which every interior elementary domain is a bigon or rectangle, fixed arcs appear only in rectangles, and triple points on the fixed set have positive Maslov contribution; their real index-1 domains are completely classified and each contributes exactly once mod 2 to the differential.

What would settle it

Exhibit a real balanced sutured manifold together with two nice real diagrams whose combinatorially computed chain complexes have non-isomorphic homology, or produce a taut real surface decomposition for which the predicted direct-sum isomorphism of real sutured Floer homology fails.

Watch

Extended reading notes

Core claim

For any real balanced sutured manifold the real sutured Floer homology is a well-defined graded vector space that can be computed from the generators and real index-1 domains of any nice real Heegaard diagram, and that transforms under taut real surface decompositions and under real arc decompositions by explicit direct-sum formulae over outer real Spin^c structures.

Load-bearing premise

The existence of generic symmetric almost-complex structures that achieve transversality for every real index-1 moduli space while remaining compatible with the anti-symplectic involution.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

0 major / 4 minor

Summary. The paper constructs real sutured Floer homology RSFH for balanced real sutured manifolds (Y, γ, τ), i.e., sutured manifolds equipped with orientation-preserving involutions whose fixed sets are properly embedded codimension-2 submanifolds. The invariant is defined by counting equivariant holomorphic curves between real intersection points of Lagrangian tori in the symmetric product (or, equivalently, in the cylindrical reformulation of Section 6), splits over relative real Spin^c structures, and is shown independent of admissible real Heegaard diagrams and generic symmetric almost-complex structures by the usual Heegaard-move and continuation-map arguments. The authors introduce real nice diagrams (Definition 8.1), prove every real balanced diagram can be made nice by real isotopies and handleslides (Theorem 8.2), classify all real index-1 domains (Proposition 8.6), and show each contributes 1 mod 2 (Propositions 8.22–8.32), yielding combinatorial computability (Theorems 1.1 and 8.33). They establish a surface-decomposition isomorphism (Theorems 1.5/9.11), an arc-decomposition formula (Theorems 1.7/5.2), real-taut hierarchies (Theorem 1.3), adjunction inequalities, and several structural differences from classical SFH (failure of Künneth, non-detection of real genus).

Significance. The work supplies a computable, diagrammatic counterpart to real monopole Floer homology and real Seiberg–Witten invariants that have already produced slice-disk and exotic-involution results. Combinatorial computability via nice real diagrams, the exhaustive classification of index-1 domains, and the surface/arc decomposition formulae are substantial technical achievements that place the theory on the same footing as Juhász’s SFH while revealing genuinely new phenomena (non-multiplicative rank under real connected sums, vanishing without real-tautness converse). The reduction of arbitrary real sutured manifolds to those without arc components of the fixed set (Remark 5.6) and the recovery of classical SFH as a special case further increase utility. These contributions are of clear interest to low-dimensional topology and Floer theory.

minor comments (4)
  1. [Introduction / throughout] Several typographical slips appear early: “equivelant” (p. 3), “Künneth Formulae” inconsistently capitalized, and occasional missing articles. A light copy-edit pass would remove them.
  2. [Sections 2, 5, 7] Figures 1–5 and 7–8 are helpful but some labels (especially the green fixed-set arcs and the purple guiding disks) become hard to distinguish in grayscale. Adding line styles or a short legend would improve accessibility.
  3. [Sections 5 and 9] The relative H_1^R-grading ε_R is introduced in Section 3.3 and used in the graded statements of Theorems 5.2 and 9.11; a one-sentence reminder of its definition at those later points would help the reader.
  4. [Sections 6 and 8] The averaging argument for transversality of immersed multiplicity-2 rectangles is only sketched (Sections 6 and 8). While the classification (Proposition 8.20, Corollary 8.21) shows these are standard forms already covered by classical arguments, a short explicit sentence confirming that the averaging step is identical to the embedded-rectangle case would remove residual ambiguity.

Circularity Check

1 steps flagged · score 1.0 of 10

No significant circularity: RSFH is defined from equivariant intersection points and real holomorphic curve counts, with invariance and combinatorial counts established by direct analysis and standard degeneration arguments rather than by construction from inputs.

  1. self citation load bearing [Proposition 2.15 / Theorem 3.5 (and surrounding text)]
    "This is proven carefully in [GM26]. … Invariance of RSFH(Y, γ, τ) essentially follows verbatim from [GM25, Section 5]."

    The relation of any two real diagrams by real Heegaard moves and the resulting quasi-isomorphisms of chain complexes are deferred to the authors’ companion notes [GM26] and the closed-case paper [GM25] (second author overlap). The present text only sketches the Morse-function path and continuation maps. This is a minor load-bearing self-citation for technical completeness, but the combinatorial core (nice diagrams, domain classification, counts) is developed independently and does not reduce to those citations by construction.

full rationale

The derivation chain is self-contained. Generators of RSFC(H) are defined as (T_α ∩ M^R) points (Section 3.1); the differential counts real index-1 equivariant strips (Eq. (1)). Invariance (Theorem 3.5) is obtained by continuation maps and real Heegaard moves, adapting the closed-case arguments of [GM25] with explicit sketches (Proposition 2.15, Lemma 3.10). Nice diagrams are constructed by equivariant finger moves and handleslides that strictly decrease a lexicographic complexity (Proposition 8.4, Theorem 8.2). Real index-1 domains are classified by non-negative contributions to the combinatorial Maslov formula (Eq. (7), Tables 1–5, Proposition 8.6); immersions are controlled (Proposition 8.20, Corollary 8.21). Holomorphic counts equal 1 mod 2 by unique branched covers for polygons (Proposition 8.23), neck-stretching/matching for reflection annuli (Proposition 8.28), conformal-parameter zeros for free-boundary annuli (Proposition 8.30), and a consistency argument using already-established invariance plus ∂^{2}=0 for antipodal annuli/tori (Propositions 8.31–8.32). Surface and arc decomposition isomorphisms follow by relating outer generators on adapted diagrams (Theorems 5.2, 9.11). The only self-citations ([GM25], [GM26]) supply technical lemmas already sketched or used as black boxes for non-real ingredients; none force the new real-sutured statements by definition. No fitted parameters, self-definitional loops, or uniqueness theorems imported circularly appear.

Assumptions & free parameters 0 free parameters · 4 assumptions · 4 invented entities

The work rests on the standard foundations of Heegaard Floer homology (symplectic forms on symmetric products, Maslov indices, transversality for almost-complex structures) together with the real-structure package of Guth–Manolescu and the sutured package of Juhász. No numerical free parameters are introduced. The new entities (real sutured manifolds, real nice diagrams, real arc decompositions) are defined by explicit geometric constructions and are therefore not free inventions; their independent evidence is the collection of theorems proved about them.

assumptions (4)
  • domain assumption Existence of generic symmetric almost-complex structures achieving transversality for real moduli spaces of index ≤1 (averaging of non-equivariant transversality).
    Invoked in §3.1, §6 and throughout §8; reduced to results of Lipshitz and Ozsváth–Szabó plus averaging, but not re-proved in full detail for immersed domains.
  • domain assumption Any two real balanced sutured Heegaard diagrams are related by a finite sequence of real Heegaard moves (stabilizations, isotopies, handleslides).
    Proposition 2.15; proved by reduction to real Morse theory following BL25 and GM26, taken as standard.
  • standard math The classical (non-real) combinatorial formula for the Maslov index and the Euler measure of domains.
    Used via the real-index formula (6) of GM25 throughout §8.
  • domain assumption Juhász’s surface-decomposition formula and the existence of nice diagrams in the non-real setting.
    The real proofs are patterned on Juh08 §§4–7; the non-real statements are black-boxed.
invented entities (4)
  • Real sutured manifold (Y,γ,τ) independent evidence
    purpose: Topological object carrying the new Floer invariant; encodes an orientation-preserving involution with codimension-2 fixed set that swaps R±.
    Definition 2.1; independent evidence is the hierarchy theorem and the recovery of closed real HF by puncturing.
  • Real nice Heegaard diagram independent evidence
    purpose: Diagrammatic model in which the RSFH differential becomes combinatorial.
    Definition 8.1; existence proved by complexity-decreasing moves; independent evidence is the explicit domain classification and mod-2 counts.
  • Real arc decomposition independent evidence
    purpose: New operation that removes fixed arcs and relates RSFH of manifolds with and without arc components of the fixed set.
    Definition 5.1 and Theorem 5.2; independent evidence is the explicit isomorphism of chain complexes after removing the unique fixed intersection point.
  • Real relative H_1-grading ϵ_R independent evidence
    purpose: Additional grading on RSFH valued in real 1-homology that is preserved by the differential and by decomposition maps.
    §3.3; independent evidence is the naturality under triangle maps and the injectivity of the map from real Spin^c structures.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Real sutured Heegaard Floer homology." pith.science (2026). https://pith.science/paper/T7VG6GN6

@misc{pith2026260711082,
  author       = {Pith},
  title        = {Pith review of: Real sutured Heegaard Floer homology},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/T7VG6GN6}},
  note         = {Machine review of arXiv:2607.11082}
}
read the original abstract

We develop a theory of real sutured manifolds and a real Heegaard Floer theory for these manifolds. We develop a notion of real nice diagrams, and prove that our invariant is combinatorially computable. Our theory shares many structural properties with Juh\'asz's sutured Floer homology, as does the topological theory of real sutured manifolds with Gabai's original sutured manifold theory. We also show that our invariant has several new structural properties differentiating it from sutured Floer homology.

Figures

Figures reproduced from arXiv: 2607.11082 by the authors.

Figure 1
Figure 1. A real sutured manifold for S 2 × [0, 1] equipped with the involution rot × id, where rot is π-rotation of S 2 . Remark 2.11. From a real (balanced) sutured Heegaard diagram (Σ, α, β, τ ), we can build a real (balanced) sutured manifold (Y, γ, τ ′ ) in which we can view Σ as an embedded real surface follows. Take the product Σ × [−1, 1] and attach 2-handles along curves αi × {−1} and βi × {1}. Set γ := ∂Σ × [−1, 1].… view at source ↗
Figure 2
Figure 2. Top: The result of a free stabilization. Bottom: The result of a fixed point stabilization. Definition 2.13. Let (H0, τ0) = (Σ0, α0, β0, τ0) and (H1, τ1) = (Σ1, α1, β1, τ1) be two real (balanced) sutured Heegaard diagrams for (Y, γ, τ ). Then we say that: (1) (H1, τ1) is obtained from (H0, τ0) by a real diffeomorphism if there exists an orientation preserving diffeomorphism φ : Y → Y isotopic to the identity such th… view at source ↗
Figure 3
Figure 3. A disk, D, guiding an arc decomposition is shown in light purple. D contains an arc component of the fixed point set Ci , shown in green. The orange curves on the right and left are arcs in the sutures. The cores of two handles in an adapted Heegaard diagram are show in red and blue. x is an intersection point that plays an important role in this section. (2) after removing ν(a) ∪ ν(∂a), there exists a properly embe… view at source ↗
Figures from the paper (32 more)
Figure 4
Figure 4. Figure 4: A neighborhood of an arc component of the fixed set Ci in a real Heegaard diagram adapted to an arc decomposition. The solid black lines are boundary components of the Heegaard surface. k indicates the multiplicity of any potential domain connecting a generator contain…
Figure 5
Figure 5. Figure 5: Using a {1}-stabilization to reduce the number of intersections with a component of the fixed set to one. Remark 5.6. The real sutured Floer homology of a real sutured manifold whose fixed point set contains exactly n arc components can be obtained by computing the rea…
Figure 6
Figure 6. Figure 6: The real sutured Heegaard diagram H0 used in analyzing the effect of adding punctures. forcing a = 2b − 1. It follows that any such domain must be of the form (2b − 1)D0 + b(D1 + D2). Similar calculations show that π R 2 (y, x) = {(2b + 1)D0 + b(D1 + D2)}b∈Z π R 2 (x, …
Figure 7
Figure 7. Figure 7: A real Heegaard diagram for S 1 × S 2 and two choices of extra punctures. c b a X X X X O1 O4 O2 O3 d f e x1 x4 x3 x2 c b a X X X O1 X O4 O2 O3 d f e x1 x4 x3 x2 q q [PITH_FULL_IMAGE:figures/full_fig_p033_7.png]
Figure 8
Figure 8. Figure 8: Real grid diagrams for complement of strongly invertible unlink and Hopf link, both with a choice of an extra puncture. 7.2. K¨unneth formulae. The usual versions of Heegaard Floer homology, knot Floer homology and sutured Floer homology satisfy K¨unneth formulae for c…
Figure 9
Figure 9. Figure 9: Left: a triple intersection with σ(α, x) = +1. Right: a triple intersection with σ(α, x) = −1. Step 4. Show that whenever the complexity of H is non-zero, we can decrease it using a finite sequence of real isotopies and handleslides. 8.1.1. Step 1. We arrange that ever…
Figure 10
Figure 10. Figure 10: An isotopy changes a neighborhood of C as shown in (A) to that shown in (B). In each subfigure the top and bottom edges should be identified. there is a marked point on C, then the lemma clearly holds. Otherwise, there is at least one pair of marked points interchange…
Figure 11
Figure 11. Figure 11: Extra finger moves when I ∩ {2, . . . , n − 1} ̸= ∅. Next assume that we are in the case of B1 (labeled as in [Juh08, Lemma 6.6]). If the finger from D′ m enters Dm from a β-edge of D1 m or D2 m, we can find an arc δ ′ with the same property as in previous case and th…
Figure 12
Figure 12. Figure 12: Extra finger moves for subcase B1 Lastly, we deal with case B2 (labeled again as in [Juh08, Lemma 6.6]). As noted above, the domains Di m, i = 1, 2, 5, 6 are already rectangles and the finger from τ (b∗) may enter D′ m from Di m for i = 1, 3, 4, 6. If the finger enter…
Figure 13
Figure 13. Figure 13: Extra finger moves for subcase B2, when the finger enters Dm from D1 m or D6 m. a1 a1 D4 m D3 m D1 m R 1 2 D2 m D′ ∗ D5 m D6 m R 2 2 R 3 2 R 4 2 a2 an ap D3 m D1 m R 1 2 D2 m D′ ∗ D4 m D5 m D6 m R 2 2 R 3 2 R 4 2 an a2 ap [PITH_FULL_IMAGE:figures/full_fig_p041_13.png]
Figure 14
Figure 14. Figure 14: Extra finger moves for subcase B2, when the finger enters Dm from D3 m or D4 m. (2) D was part of Dm (or D′ m) after the finger move. By construction, D has distance less than d0, since it is still adjacent to D∗. (3) D was some Dl (or D′ l ) with l ̸= m with badness …
Figure 15
Figure 15. Figure 15: Possible configurations of multiplicities at vertices on C organized by their contribution to V{1} . For example, the first frame shows a domain with multiplicity 1/4 and triple Maslov index 1/2 at an outgoing vertex y, which therefore contributes 1 4 + 1 2 to V{1} . …
Figure 16
Figure 16. Figure 16: Symmetric domains which may appear in a nice diagram with no vertices on C. The figure also depicts tilings of the annular domains into rectangles. The left pictures in the second and third row are for case (c), while the right ones are for (d). In the lower rows, the…
Figure 17
Figure 17. Figure 17: Symmetric disk domains which may appear in a nice diagram with at least two vertices on C. These five cases correspond to the following domains: (1) F is a disk with a rectangular boundary. Its boundary has two 90◦ corners interchanged by the reflection and two 90◦ co…
Figure 18
Figure 18. Figure 18: A symmetric disk domain which cannot appear in a nice diagram. with multiplicities), thus there are exactly two bigons appear in the tiling of F. See Case (2) in [PITH_FULL_IMAGE:figures/full_fig_p051_18.png]
Figure 19
Figure 19. Figure 19: Symmetric annular domains which may appear in a nice diagram with at least two vertices on C. (8) F may have two rectangle boundaries or one bigon boundary and one hexagon boundary. In the former case, the two boundary components each have a pair of 90◦ corners interc…
Figure 20
Figure 20. Figure 20: A symmetric toroidal domain which may appear in nice diagrams with at least two vertices on C. In the real setting (and in contrast to the usual one), it is possible that uΣ : F → Σ fails to be an embedding, i.e. we may witness domains which are merely immersed in Σ, …
Figure 21
Figure 21. Figure 21: In the top left frame, we show the four possible start paths which can be taken by a. The remaining frames show the minimal completions of these paths in the cases indicated. 8.3.3. Analysis of immersions. We now state the possible immersions in a real nice diagram us…
Figure 22
Figure 22. Figure 22: A minimal completion of the path a˜. The β-curve shown must create a bigon, which is prohibited. Proposition 8.20. Let D be any of the real index 1 domains from Proposition 8.6. If D is not embedded, then the induced tiling is simple. Moreover, if D = P i aiDi is writ…
Figure 23
Figure 23. Figure 23: A tiling of the octagon on the left and a possible immersion on the right. In [PITH_FULL_IMAGE:figures/full_fig_p058_23.png]
Figure 24
Figure 24. Figure 24: A tiling on the (2,4)-annulus and a possible immersion of such a domain. Free-boundary reflection annuli: A similar argument as above will show that the domain of a free-boundary reflection annulus can never be immersed. More precisely, in [PITH_FULL_IMAGE:figures/fu…
Figure 25
Figure 25. Figure 25: A tiling on a free-boundary reflection-annulus and a possible immersion of such a domain. p1 p2 q1 q3 q2 a b c [PITH_FULL_IMAGE:figures/full_fig_p060_25.png]
Figure 26
Figure 26. Figure 26: Left: A tiling on the toroidal domain. Right: A possible immersion of a toroidal domain; here, the two turquoise rectangles are identified. Corollary 8.21. Suppose Pk i=1 miDi is the domain of a real pseudo-holomorphic ϕ ∈ π2(x, y) with µR(ϕ) = 1. If mi > 1 then Di is…
Figure 27
Figure 27. Figure 27: Each frame consists of two pictures; the left one shows a tiling structure in which all possible choices of p and q, the right one shows an example of immersion of that structure in a real Heegaard diagram. Top left: bigon; Top right: hexagon; Middle left: (2,2)-bound…
Figure 28
Figure 28. Figure 28: A real polygon. Proof. Let H denote the upper half plane in C and let D denote the standard unit disk in C. The formula a + ib 7→ −a + ib gives real structures c on H and D. Note that the function f : H → D defined by z 7→ −i z − i z + i is an equivariant holomorphic …
Figure 29
Figure 29. Figure 29: Breaking an annulus into a pair of polygons via neck stretching. are real polygons which satisfy a matching condition. Both moduli spaces are 1-dimensional by Proposition 8.23, parametrized by the position of the marked point. Hence, McR(D1) ×ρ McR(D2) is 1-dimensiona…
Figure 30
Figure 30. Figure 30: A 1-parameter family of holomorphic curves from ac to cd. Proposition 8.30. Let F be a free-boundary reflection annulus from Section 8.2.4, and let D be a corresponding domain in Σ. For any adapted and generic path of symmetric almost complex structures J, we have tha…
Figure 31
Figure 31. Figure 31: Top: cuts on a Z2-reflection annulus. Bottom: a degenerated annulus and its identification with the standard disk. Finally, we deal with the antipodal annuli from Section 8.2.4 and the tori from Section 8.2.8. Each has an odd number of symmetric holomorphic representa…
Figure 32
Figure 32. Figure 32: A real Heegaard diagram for a real lens space containing toroidal and antipodal annular domains. Proof. As in the proof of Proposition 8.28, we can reduce the problem of analyzing the moduli space from the same source to a specific nice real Heegaard diagram. Consider…
Figure 33
Figure 33. Figure 33: The projection map from D(P) — shown on the left — to a real Heegaard diagram adapted to P. Note that the green axis in the left hand diagram is not contained in the 3-manifold, and is for illustrative purposes only. This same is trur for the upper and lower portions …
Figure 34
Figure 34. Figure 34: Model cases of immersions of real domains in a surface diagram. h : F → Σ ′ by h(x) :=    p −1 (x) if u(x) ∈ D \ P p −1 (x) ∩ PA if u(x) ∈ DA p −1 (x) ∩ PB if u(x) ∈ DB. It can be checked that h is continuous map as in the proof of [Juh08, Proposition 7.6]. Since…
Figure 35
Figure 35. Figure 35: The knot 83 is pictured on the left. A minimal genus equivariant Seifert surface for 83 is shown on the right, together the boundaries of a pair of compressing disks for the Seifert surface. The vertical lines are the axes of the standard involution of S 3 . involutio…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

14 extracted references · 9 linked inside Pith

  1. [1]

    Morse homology and equivariance.arXiv preprint arXiv:2409.04694,

    [BL25] Erkao Bao and Tyler Lawson. Morse homology and equivariance.arXiv preprint arXiv:2409.04694,

  2. [2]

    Real Heegaard Floer homology.arXiv preprint arXiv:2504.09034,

    [GM25] Gary Guth and Ciprian Manolescu. Real Heegaard Floer homology.arXiv preprint arXiv:2504.09034,

  3. [3]

    A note on real Heegaard Floer homology and localization.arXiv preprint arXiv:2508.03897,

    [Hen25] Kristen Hendricks. A note on real Heegaard Floer homology and localization.arXiv preprint arXiv:2508.03897,

  4. [4]

    Invariant Seifert surfaces for strongly invertible knots

    [HHS23] Mikami Hirasawa, Ryota Hiura, and Makoto Sakuma. Invariant Seifert surfaces for strongly invertible knots. 34:325–349, [2023]©2023. [HKM25] Mark Hughes, Seungwon Kim, and Maggie Miller. Branched covers of twist-roll spun knots and turned twisted tori

  5. [5]

    [Juh06] Andr´ as Juh´ asz

    Publisher Copyright:©2021 American Mathematical Society. [Juh06] Andr´ as Juh´ asz. Holomorphic discs and sutured manifolds.Algebraic & Geometric Topology, 6(3):1429–1457,

  6. [6]

    Smooth concordance of cables of the figure-eight knot.arXiv preprint arXiv:2505.03720,

    [KPT25] Sungkyung Kang, JungHwan Park, and Masaki Taniguchi. Smooth concordance of cables of the figure-eight knot.arXiv preprint arXiv:2505.03720,

  7. [7]

    Monopole Floer homology and real structures.arXiv preprint arXiv:2211.10768,

    [Li22] Jiakai Li. Monopole Floer homology and real structures.arXiv preprint arXiv:2211.10768,

  8. [8]

    Multi-framed real monopole floer theory.arXiv preprint arXiv:2606.28596,

    [Li26] Jiakai Li. Multi-framed real monopole floer theory.arXiv preprint arXiv:2606.28596,

Show all 14 references
  1. [9]

    Real bordered Floer homology.arXiv preprint arXiv:2604.20565,

    [LO26] Robert Lipshitz and Peter Ozsv´ ath. Real bordered Floer homology.arXiv preprint arXiv:2604.20565,

  2. [10]

    A gauge theoretic invariant of embedded surfaces in 4-manifolds and exotic P 2-knots.arXiv preprint arXiv:2312.02041,

    [Miy23] Jin Miyazawa. A gauge theoretic invariant of embedded surfaces in 4-manifolds and exotic P 2-knots.arXiv preprint arXiv:2312.02041,

  3. [11]

    A satellite formula for real Seiberg-Witten Floer homotopy types.arXiv preprint arXiv:2504.03270,

    [MPT25] Jin Miyazawa, JungHwan Park, and Masaki Taniguchi. A satellite formula for real Seiberg-Witten Floer homotopy types.arXiv preprint arXiv:2504.03270,

  4. [12]

    [OSS] Peter Ozsv´ ath, Zolt´ an Szab´ o, and Andr´ as I

    Publisher: Mathematical Sciences Publishers. [OSS] Peter Ozsv´ ath, Zolt´ an Szab´ o, and Andr´ as I. Stipsicz. Heegaard Floer homology.https: // web. math. princeton. edu/ ~ petero/ konyv7. pdf. [Per08] Timothy Perutz. Hamiltonian handleslides for Heegaard Floer homology. InP...

  5. [13]

    Real link Floer homology.arXiv preprint arXiv:2604.21240,

    [Xia26] Yonghan Xiao. Real link Floer homology.arXiv preprint arXiv:2604.21240,

  6. [14]

    Graph cobordisms and Heegaard Floer homology.arXiv preprint arXiv:1512.01184,

    97 [Zem19] Ian Zemke. Graph cobordisms and Heegaard Floer homology.arXiv preprint arXiv:1512.01184,

Pith tools

Reviewed July 14, 2026 · model on record in the stance chip above.