Moduli spaces of spacefilling branes in symplectic 4-manifolds

Charlotte Kirchhoff-Lukat; Marco Zambon

REVIEW 2 major objections 5 minor 1 cited by

Moduli spaces of spacefilling branes in symplectic 4-manifolds

T0 review · 2 major / 5 minor · reviewed 2026-08-10 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read The moduli space of spacefilling brane structures on K3 surfaces and 4-tori is a non-Hausdorff smooth manifold of dimension 20 and 4, with a local diffeomorphism to an explicit cohomology quadric.

desk verdict Solid K3 result, but the 4-torus half of the main theorem is wrong: M_ω is Hausdorff, not non-Hausdorff. read the letter →

arxiv 2501.07926 v2 pith:FT4ASZEG submitted 2025-01-14 math.SG math.AGmath.DG

Charlotte Kirchhoff-Lukat , Marco Zambon This is my paper · ORCID

classification math.SGmath.AGmath.DG MSC 53D0514J2832G13

keywords spacefillingbranessymplectic4-manifoldsmodulispacesK3surfacescomplextorilocalTorellitheoremholomorphicformsperiodmap

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 establishes a global deformation statement for spacefilling branes on the two compact Kähler 4-manifolds that admit holomorphic symplectic forms: K3 surfaces and 4-tori. Fixing any symplectic form that admits at least one such brane, the space of all brane structures modulo cohomologically trivial symplectomorphisms is a smooth manifold, non-Hausdorff and non-compact, of real dimension 20 for K3 and 4 for the torus. The map sending a brane to its cohomology class is a local diffeomorphism onto the quadric cut out by $[F']\wedge[\omega]=0$ and $[F']\wedge[F']=[\omega]\wedge[\omega]$. This matters because it reduces a nonlinear, infinite-dimensional deformation problem to ordinary cohomology: nearby brane structures up to equivalence are exactly nearby points of an explicit quadric. The proof relies on the classical local Torelli theorem for K3 surfaces and tori.

What carries the argument

The restricted period map. A brane $F$ corresponds to the complex structure $I=\omega^{-1}\circ F$, for which $F+i\omega$ is holomorphic symplectic; in real dimension 4, complex structures with holomorphic symplectic forms are in bijection with complex lines of closed 2-forms satisfying $[\Omega]\wedge[\Omega]=0$ and $[\Omega]\wedge\overline{[\Omega]}>0$ (Proposition 3.4). Restricting to lines whose imaginary part is $\omega$ identifies brane structures with points of the real quadric $\mathcal{Q}_{[\omega]}$. The local Torelli theorem makes this restricted period map a local diffeomorphism modulo cohomologically trivial diffeomorphisms, and Lemma 4.14 shows the relevant quotient can be taken by symplectomorphisms rather than general diffeomorphisms. The load-bearing comparison is therefore between $\mathcal{M}_\omega$ and the quadric $\mathcal{Q}_{[\omega]}$, with the period map providing the local chart.

What would settle it

A concrete falsifier is a single brane $F$ on the K3 manifold or $T^4$ where the derivative of the map $\Phi$ is not an isomorphism onto the tangent space of $\mathcal{Q}_{[\omega]}$, for instance a closed real $(1,1)$-form $\alpha$ satisfying the linearized equations that cannot be integrated to an actual nearby brane. The 4-torus case is explicitly checkable in the coordinates of Example 4.23; a failure there, or on K3 at a non-standard complex structure, would refute the theorem.

Watch

Extended reading notes

Core claim

On the K3 manifold or the 4-torus $M$, for any symplectic form $\omega$ admitting a spacefilling brane, the moduli space $\mathcal{M}_\omega$ of spacefilling brane structures modulo symplectomorphisms acting trivially on cohomology is a non-Hausdorff smooth manifold. The map $\Phi([F])=[F]$ is smooth and a local diffeomorphism onto the quadric $\mathcal{Q}_{[\omega]}\subset H^2(M,\mathbb{R})$ defined by $[F']\wedge[\omega]=0$ and $[F']\wedge[F']=[\omega]\wedge[\omega]$. Consequently the moduli space is locally modelled on $\mathcal{Q}_{[\omega]}$; since $\mathcal{Q}_{[\omega]}\simeq S^1\times\mathbb{R}^{b_2-3}$, the dimension is $b_2-2$, namely 20 for K3 ($b_2=22$) and 4 for the torus ($b_2=6$), and the moduli space is non-compact. The theorem applies to every symplectic form admitting a brane because every complex structure on these manifolds is Kähler.

Load-bearing premise

The load-bearing premise is that the local Torelli rigidity theorem holds at every relevant complex structure and that, near each point of the period domain, a smooth universal family of deformations exists: small deformations of the complex structure are then faithfully recorded, up to diffeomorphisms that leave all cohomology classes unchanged, by the cohomology class of their holomorphic 2-form.

Editorial extensions

If this is right

For every brane $F$, a neighborhood of its class in $\mathcal{M}_\omega$ is diffeomorphic to a neighborhood of $[F]$ in $\mathcal{Q}_{[\omega]}$; every sufficiently small cohomology class satisfying the linearized equations is therefore realized by an actual brane up to equivalence.
The moduli space is non-Hausdorff: the non-separation of nearby complex structures in the period domain carries over, so distinct brane structures can be inseparable by open sets.
The tangent space to $\mathcal{M}_\omega$ at a brane is $H^{1,1}_{\mathbb{R}}(M)$ for the associated complex structure, refining the infinitesimal statement that deformations are closed real $(1,1)$-forms.
The quadric $\mathcal{Q}_{[\omega]}$ is diffeomorphic to $S^1\times\mathbb{R}^{b_2-3}$ and carries a Lorentzian metric of signature $(1,b_2-3)$; through the local diffeomorphism these give a local model and a causal structure for the moduli space.
The moduli space is non-compact, because its image in $\mathcal{Q}_{[\omega]}$ is open and non-compact while any compact image would be closed and hence all of the connected quadric.

Reading between the lines

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

For the 4-torus the paper proves $\Phi$ is surjective; the same is left open for K3. A natural testable conjecture is surjectivity on K3 as well, since the period map is globally surjective and the remaining step would be to control the diffeomorphism that changes the symplectic form within its cohomology class.
The Lorentzian structure on $\mathcal{Q}_{[\omega]}$ suggests a causal splitting of brane deformations into one positive and $b_2-3$ negative directions; whether this has a direct string-theoretic meaning is not addressed in the paper.
The same local-Torelli strategy would apply to compact hyperkähler manifolds of higher dimension, where a local Torelli theorem is known; the paper explicitly does not pursue this, and the missing ingredient would be a higher-dimensional analogue of the dimension-4 correspondence between complex structures and lines of 2-forms.

Share X Bluesky LinkedIn Reddit HN

Formalized claims in Lean

Claim #1: On the K3 manifold or the 4-torus $M$, for any symplectic form $\omega$ admitting a spacefilling brane, the moduli space $\mathcal{M}_\omega$ of spacefilling brane structures modulo symplectomorphisms acting trivially on cohomology is a non-Hausdorff smooth manifold. The map $\Phi([F])=[F]$ is smooth and a local diffeomorphism onto the quadric $\mathcal{Q}_{[\omega]}\subset H^2(M,\mathbb{R})$ defi

/-- @claim 1 On the K3 manifold or the 4-torus $M$, for any symplectic form $\omega$ admitting a spacefilling brane, the moduli space $\mathcal{M}_\omega$ of spacefilling brane structures modulo symplectomorphisms acting trivially on cohomology is a non-Hausdorff smooth manifold. The map $\Phi([F])=[F]$ is smooth and a local diffeomorphism onto the quadric $\mathcal{Q}_{[\omega]}\subset H^2(M,\mathbb{R})$ defi -/
def central_claim : Prop :=

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.

Desk Editor's Note

read the letter

Here's my take on arXiv:2501.07926. The genuinely new content is the global description of the moduli space of spacefilling branes on K3 and T4: smoothness, dimension (20 and 4), local diffeomorphism to the real quadric Q_[ω], plus the deformation equations and the explicit torus computation. The proofs of the new steps—restricting the period map, transversality of the fixed-imaginary-part slice, Lemma 4.14 comparing Diff^* and Symp^*, and the quadric analysis—are careful and mostly self-contained, with external reliance only on standard Local Torelli and smooth universal family results. I have no circularity concerns.

But the paper's blanket non-Hausdorff claim for T4 is false. The argument for non-Hausdorffness comes from Remark 3.8, which states that the domain of the period map P is non-Hausdorff. That is true for the quotient by the full mapping class group, not for the quotient by Diff^*(M) used in the paper. For T4, every complex structure is Diff^*-equivalent to a constant one; for a constant complex structure the period line is the Plücker line of the Hodge plane H^{1,0}⊂H^1(T4,C), and the Plücker embedding of Gr(2,4) is injective. So P is injective on the Diff^* quotient. Consequently P_ω and Φ are injective local diffeomorphisms onto Q_[ω], and M_ω is Hausdorff. In fact, using the paper's own Proposition 4.17 (surjectivity of Φ), M_ω for T4 is diffeomorphic to Q_[ω] ≅ S^1×R^3. So Theorem 4.15 and Corollary 4.22 are wrong in the T4 half. The K3 half may well be correct: non-Hausdorffness is a real feature of the K3 moduli space, and the rest of the argument does not depend on the torus case.

One small gap in Proposition 4.17: the proof asserts a diffeomorphism making ω constant, but the reduction to the constant case requires that diffeomorphism to lie in Diff^* (or at least preserve cohomology classes). This is fixable, since the diffeomorphism can be chosen homotopic to a linear identification, hence in Diff_0.

Bottom line: the local moduli description and the K3 global statements are worth having, and the paper should go to a referee. But the T4 non-Hausdorffness needs to be corrected before publication; as written, a central stated theorem is false for one of the two main cases.

Referee Report

2 major / 5 minor

Summary. The paper studies spacefilling brane structures on a symplectic 4-manifold (M,ω): closed 2-forms F such that ω^{-1}∘F is an integrable complex structure with F+iω holomorphic symplectic. For M a K3 surface or the 4-torus and ω any symplectic form admitting such a brane, the authors define the moduli space M_ω as the quotient of brane structures by Symp^*(M,ω), and claim that M_ω is a non-Hausdorff smooth manifold of dimension 20 (K3) or 4 (T^4), that it is non-compact, and that the cohomology class map Φ:M_ω→Q_[ω] is a smooth local diffeomorphism onto an explicit real quadric. The proof restricts the classical period map to complex structures for which ω(I·,·) is skew, applies the local Torelli theorem, and identifies the restricted period map with Φ via a lemma comparing Diff^* and Symp^* equivalences.

Significance. If the main theorem is correct, the paper gives a global description of the moduli space of spacefilling branes, identifying it locally with a simple real quadric and computing its dimension. The restriction of the period map to the fixed-imaginary-part slice, the explicit treatment of the quadric Q_[ω], and the comparison of Diff^* and Symp^* equivalences in Lemma 4.14 are useful and original contributions. The reliance on the local Torelli theorem is clearly stated and is a standard external input. However, the non-Hausdorff assertion is false for the 4-torus, as explained below, and this is a central claim of the paper's main theorem.

major comments (2)

[§4.3.2, Theorem 4.15; §3.2, Remark 3.8] The asserted non-Hausdorffness is false for the 4-torus. For T^4 every complex structure is Diff^*-equivalent to a translation-invariant one, as the paper itself establishes in Proposition 4.17. For a translation-invariant complex structure, the period line C[Ω] is the Plücker line of the Hodge plane H^{1,0}⊂H^1(T^4,C); the Plücker embedding Gr(2,4)→P(∧^2 C^4) is injective, so the period line determines H^{1,0} and hence the complex structure. Therefore the period map P of (6) is injective on the Diff^* quotient. Lemma 4.14 identifies M_ω with the domain of the restricted period map P_ω (and its proof actually works without the 'nearby' assumption), while Proposition 4.17 shows P_ω is surjective onto Q_[ω]. Consequently Φ is an injective local diffeomorphism onto the Hausdorff space Q_[ω], so M_ω is Hausdorff, contradicting Theorem 4.15. The statement in Remark 3.8 that the domain of P is non-Hausdorff is true only for the quotient by the full mapping class group, not for the Diff^* quotient used in this paper.
[§4.3.2, Theorem 4.15] Even if the intended non-Hausdorff claim is restricted to K3 surfaces, the paper does not prove it. Theorem 4.15 asserts non-Hausdorffness without argument; the only support is Remark 3.8, which is an unproved 'well-known' statement and is false for T^4. If the K3 assertion is retained, the authors should supply either a proof or a precise reference establishing non-Hausdorffness of the moduli space of complex structures modulo Diff^*(M), not modulo the full mapping class group.

minor comments (5)

[§1, p.2] The phrase 'antisymmetric 2' is incomplete; it should read 'antisymmetric 2-form'.
[§4.3.2, Lemma 4.14] The lemma is stated for 'nearby' brane structures, but the proof does not use proximity and works globally. Stating it without the proximity assumption would make the identification of M_ω with the domain of P_ω cleaner and would avoid confusion about the global equivalence of the two quotients.
[§3.2, Remark 3.8] The sentence 'It is well-known that the domain of the period map P is non-Hausdorff' needs revision and a citation; as it stands it is false for the 4-torus case considered in this paper.
[§4.4, Proposition 4.19] The sentence 'Pick an element b∈H^{1,1}_R with b∧b=[ω]∧[ω]' would benefit from a brief justification of the existence of such a b, since H^{1,1}_R has signature (1,b_2-3).
[Appendix A, Corollary A.3] 'de Rham cohohomology' is a typo for 'de Rham cohomology'.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the brane moduli result follows from the external Local Torelli theorem plus independent quadric computations.

full rationale

The claimed derivation is not circular. The main result (Theorem 4.15 and Corollary 4.22) is obtained by composing three independent inputs: (1) the elementary equivalence of a spacefilling brane F with a complex structure I for which ω(I·,·) is skew and F+iω is holomorphic symplectic (Proposition 2.6, Lemma 2.5); (2) the standard identification of complex structures with lines CΩ satisfying HS(1)–(3) (Proposition 3.4), restricted to the slice Im[Ω']=[ω], which yields the diffeomorphism between {C[Ω']∈Q: Im[Ω']=[ω]} and the real quadric Q_[ω] (Lemmas 4.8, 4.10, 4.13); and (3) the Local Torelli Theorem 3.7, quoted from Huybrechts and Manetti/Huybrechts, which is an external classical theorem and is not a restatement of the paper's conclusion. The identification of the Diff*-quotient with the Symp*-moduli space in Lemma 4.14 is proved from Lemma 4.8 and the scalar uniqueness of holomorphic symplectic forms; it is not built into the definition of M_ω. The dimension and non-compactness statements are consequences of the explicit cylinder description Q_[ω]≅S^1×R^{b2−3} (Propositions 4.19, 4.21), a computation independent of the period map. The only self-citation is the companion paper [14], referenced in Remark 4.20 for future prequantised branes; it is not load-bearing. The skeptic's objection that M_ω for the 4-torus may be Hausdorff would be a substantive mathematical error in the paper, but it is not a circularity of the sort defined here: no output is equivalent to an input by construction, and no load-bearing step reduces to a self-citation.

Assumptions & free parameters 0 free parameters · 8 assumptions · 0 invented entities

The theorem is built on classical classification, Torelli, Moser, and Hodge theory results. No parameters are fitted to data and no new entities are postulated. The only hypotheses are the existence of a brane and Kählerity, which is automatic for the two manifold classes considered.

assumptions (8)

standard math Newlander-Nirenberg theorem: an almost complex structure is integrable iff its Nijenhuis tensor vanishes.
Used in the proof of Lemma 2.5 to show that dF=0 is equivalent to integrability of I=omega^{-1} composed with F, and therefore in Proposition 2.6.
standard math Local Torelli theorem for compact Kähler 4-manifolds with holomorphic symplectic form: the period map P is a local diffeomorphism onto the period domain.
Theorem 3.7 is the engine for Theorem 4.15. It is cited with a provenance discussion and is not proved in the paper.
standard math Kodaira classification of compact complex surfaces: trivial canonical bundle occurs only for K3 surfaces, complex tori, and primary Kodaira surfaces, with K3 and tori Kähler.
Used in Proposition 3.4 and Remark 3.6 to restrict the underlying smooth manifold to the K3 manifold or the 4-torus.
standard math Todorov-Siu theorem: every K3 surface is Kähler.
Used in Remark 3.6 and after Theorem 3.7 to ensure that locally Torelli applies at every complex structure on the K3 manifold.
standard math Moser's theorem for cohomologous symplectic forms on compact manifolds.
Used in Lemma 4.9 to find a diffeomorphism in Diff^*(M) pulling back omega' to omega, and in Proposition 4.17 for the torus.
standard math Existence of smooth universal families of deformations over an open neighborhood of the period domain for K3 surfaces and complex tori.
Invoked in Lemma 4.9 and cited to Huybrechts for K3 surfaces and Manetti for tori. If this failed, the local bijection between brane deformations and period lines would lack a proof.
standard math Surjectivity of the period map P for K3 surfaces and 4-tori.
Theorem 3.9 is used for Remark 4.18 about the image of Phi. It is not needed for the main smoothness and dimension theorem, but it is a stated global property.
standard math The d-bar d-bar lemma and Hodge decomposition for compact Kähler manifolds.
Used in Appendix A to identify the real de Rham and Dolbeault cohomology splittings that underlie Remark 4.16 and Proposition 4.19.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Moduli spaces of spacefilling branes in symplectic 4-manifolds." pith.science (2026). https://pith.science/paper/FT4ASZEG

@misc{pith2026250107926,
  author       = {Pith},
  title        = {Pith review of: Moduli spaces of spacefilling branes in symplectic 4-manifolds},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/FT4ASZEG}},
  note         = {Machine review of arXiv:2501.07926}
}

abstract

On a symplectic manifold $(M, \omega)$, a spacefilling brane structure is a closed 2-form $F$ which determines a complex structure, with respect to which $F +i\omega$ is holomorphic symplectic. For holomorphic symplectic compact K\"ahler 4-manifolds, we show that the moduli space of spacefilling branes is smooth, and determine its dimension. The proof relies on the local Torelli theorem for K3 surfaces and tori.

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

Coisotropic branes in symplectic manifolds
math.SG 2025-07 conditional novelty 7.0 of 10

The authors develop a deformation theory for coisotropic branes, explicitly describing nearby branes in a codimension-one model and showing that the forgetful map from branes to coisotropic submanifolds is not locally...

Reference graph

Works this paper leans on

22 extracted references · 20 canonical work pages · cited by 1 Pith paper

[1]

Beauville

A. Beauville. Vari´ et´ es k¨ ahleriennes dont la premi` ere classe de Chern est nulle.J. Differ. Geom., 18:755–782, 1983

work page 1983
[2]

Bischoff and M

F. Bischoff and M. Gualtieri. Brane quantization of toric Poisson varieties.Com- munications in Mathematical Physics, 391, Aug. 2022. 23

work page 2022
[3]

Bogomolov, N

F. Bogomolov, N. Kurnosov, A. Kuznetsova, and E. Yasinsky. Geometry and auto- morphisms of non-K¨ ahler holomorphic symplectic manifolds.International Mathe- matics Research Notices, 2022(16):12302–12341, Aug. 2022

work page 2022
[4]

Buchdahl

N. Buchdahl. Compact K¨ ahler surfaces with trivial canonical bundle.Ann. Global Anal. Geom., 23(2):189–204, 2003

work page 2003
[5]

Gualtieri.Generalized complex geometry

M. Gualtieri.Generalized complex geometry. PhD thesis, University of Oxford, 2003, Preprint ArXiv:math/0401221

arXiv 2003
[6]

Gualtieri

M. Gualtieri. Branes on Poisson varieties. InThe many facets of geometry, pages 368–394. Oxford Univ. Press, Oxford, 2010

work page 2010
[7]

Huybrechts

D. Huybrechts. Geometries on K3 surfaces and their symmetries. Available athttp: //www.math.uni-bonn.de/people/huybrech/Publnotes.html?language=en

work page
[8]

Huybrechts.Complex geometry

D. Huybrechts.Complex geometry. Universitext. Springer-Verlag, Berlin, 2005

work page 2005

Show all 22 references

[9]

Huybrechts

D. Huybrechts. Generalized Calabi-Yau structures,K3 surfaces, andB-fields.In- ternat. J. Math., 16(1):13–36, 2005

2005
[10]

Huybrechts

D. Huybrechts. A global Torelli theorem for hyperk¨ ahler manifolds. InS´ eminaire Bourbaki. Volume 2010/2011. Expos´ es 1027–1042. Avec table par noms d’auteurs de 1948/49 ` a 2009/10, pages 375–403. Paris: Soci´ et´ e Math´ ematique de France (SMF), 2012

2010
[11]

Huybrechts.Lectures onK3 surfaces, volume 158

D. Huybrechts.Lectures onK3 surfaces, volume 158. Cambridge University Press, 2016

2016
[12]

Kapustin and Y

A. Kapustin and Y. Li. Open-string BRST cohomology for generalized complex branes.Advances in Theoretical and Mathematical Physics, 9(4):559–574, 2005

2005
[13]

Kapustin and D

A. Kapustin and D. Orlov. Remarks on A-branes, mirror symmetry, and the Fukaya category.J. Geom. Phys., 48(1):84–99, 2003

2003
[14]

Kirchhoff-Lukat and M

C. Kirchhoff-Lukat and M. Zambon. Prequantised spacefilling branes in symplectic manifolds. In preparation
[15]

Lieberman, R

D. Lieberman, R. Wilsker, and C. Peters. A theorem of local-Torelli type.Mathe- matische Annalen, 231(1):39–45, Feb. 1977

1977
[16]

M. Manetti. Lectures on deformations of complex manifolds (deformations from differential graded viewpoint).Rend. Mat. Appl. (7), 24(1):1–183, 2004

2004
[17]

Moroianu.Lectures on K¨ ahler Geometry

A. Moroianu.Lectures on K¨ ahler Geometry. London Mathematical Society Student Texts. Cambridge University Press, 2007

2007
[18]

Oh and J.-S

Y.-G. Oh and J.-S. Park. Deformations of coisotropic submanifolds and strong homotopy Lie algebroids.Invent. Math., 161(2):287–360, 2005

2005
[19]

Y. Siu. Every K3 surface is K¨ ahler.Inventiones Mathematicae, pages 139–150, 1983

1983
[20]

Soldatenkov and M

A. Soldatenkov and M. Verbitsky. The Moser isotopy for holomorphic symplectic and C-symplectic structures, 09 2021, Preprint ArXiv:math/02109.00935. 24

2021 arXiv
[21]

A. N. Todorov. Applications of the K¨ ahler-Einstein-Calabi-Yau Metric to Moduli of K3 Surfaces.Inventiones Mathematicae, pages 251–266, 1980

1980
[22]

A. Weil. Final report on contract AF18(603)-57. (In Collected Works). 25

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.

[1] [1]

Beauville

A. Beauville. Vari´ et´ es k¨ ahleriennes dont la premi` ere classe de Chern est nulle.J. Differ. Geom., 18:755–782, 1983

work page 1983

[2] [2]

Bischoff and M

F. Bischoff and M. Gualtieri. Brane quantization of toric Poisson varieties.Com- munications in Mathematical Physics, 391, Aug. 2022. 23

work page 2022

[3] [3]

Bogomolov, N

F. Bogomolov, N. Kurnosov, A. Kuznetsova, and E. Yasinsky. Geometry and auto- morphisms of non-K¨ ahler holomorphic symplectic manifolds.International Mathe- matics Research Notices, 2022(16):12302–12341, Aug. 2022

work page 2022

[4] [4]

Buchdahl

N. Buchdahl. Compact K¨ ahler surfaces with trivial canonical bundle.Ann. Global Anal. Geom., 23(2):189–204, 2003

work page 2003

[5] [5]

Gualtieri.Generalized complex geometry

M. Gualtieri.Generalized complex geometry. PhD thesis, University of Oxford, 2003, Preprint ArXiv:math/0401221

arXiv 2003

[6] [6]

Gualtieri

M. Gualtieri. Branes on Poisson varieties. InThe many facets of geometry, pages 368–394. Oxford Univ. Press, Oxford, 2010

work page 2010

[7] [7]

Huybrechts

D. Huybrechts. Geometries on K3 surfaces and their symmetries. Available athttp: //www.math.uni-bonn.de/people/huybrech/Publnotes.html?language=en

work page

[8] [8]

Huybrechts.Complex geometry

D. Huybrechts.Complex geometry. Universitext. Springer-Verlag, Berlin, 2005

work page 2005