Pith. sign in

REVIEW 2 major objections 3 minor 40 references

A smooth family of $G_2$-instantons over a generalised Kummer construction

T0 review · 2 major / 3 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read This paper constructs the first smooth one-parameter family of G₂-instantons over a compact G₂-manifold, extending the gluing method of [Wal13a] and using a ℤ₂ symmetry of the orbifold to cancel the obstruction that would force instantons…

desk verdict A likely first smooth family of G2-instantons, but one sketched estimate needs to be completed before the result is fully established. read the letter →

arxiv 2507.00655 v1 pith:NIJSQGU5 submitted 2025-07-01 math.DG

Dominik Gutwein This is my paper · ORCID
classification math.DG MSC 53C0753C2553C29
keywords G₂-instantonsgeneralisedKummerconstructionG₂-orbifoldsgluingHermitianYang–MillsconnectionsALECalabi–Yaumanifoldsmodulispaceofinstantonsinfinitesimalrigidity
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

$G_2$-instantons — connections whose curvature satisfies $F_A \wedge \psi = 0$ for the coassociative 4-form $\psi$ — are the Yang–Mills minima in seven dimensions, and a standing question is what their moduli spaces look like. This paper claims that on a compact $G_2$-manifold produced by Joyce's generalised Kummer construction, instantons need not be isolated: for each sufficiently small gluing scale it produces a whole smooth interval of distinct instantons on a single $\operatorname{SO}(14)$-bundle, all non-flat and infinitesimally irreducible, mapping injectively into the moduli space modulo gauge. The construction extends the gluing method of [Wal13a] to families of connections whose linearised instanton operator has a non-trivial cokernel, and then uses a $\mathbb{Z}_2$-symmetry of the underlying orbifold (orientation reversal $y \mapsto -y$) so that only the invariant part of the cokernel has to vanish. The author notes that this is to his knowledge the first example of a smooth one-parameter family of instantons over a compact $G_2$-manifold, the kind of structure that any future enumerative theory of $G_2$-instantons would have to accommodate.

What carries the argument

The load-bearing object is the linearised instanton operator $L_A = (\psi_t \wedge d_A - d^*_A,\, d^*_A)$ together with its formal adjoint $L^*_A$, acting on weighted Hölder spaces adapted to the Kummer degeneration; the proof runs a fixed-point contraction for the augmented instanton equation $F_A \wedge \psi - d^*_A \xi = 0$ using the right-inverse $L^*_{\tilde A}(L_{\tilde A}L^*_{\tilde A})^{-1}$, and everything depends on a scale-uniform estimate for $L_{\tilde A}L^*_{\tilde A}$ on $\mathbb{Z}_2$-invariant sections, the t-uniform linear estimate of Proposition 5.16. Two named inputs carry the example: the representation family $f_\theta : \Gamma \to \operatorname{SO}_{\mathbb{R}}(\mathbb{C}^7)$ of the crystallographic group $\Gamma$, giving a one-parameter family of flat bundles whose infinitesimal direction is the kernel of the linearised operator, and the model operator $L_{\hat A + t\hat a}L^*_{\hat A + t\hat a}$ on $\mathbb{R} \times \hat Z_\zeta$, the ALE Calabi–Yau resolution of $\mathbb{C}^3/\mathbb{Z}_7$, whose vanishing $\mathbb{Z}_2$-invariant cokernel (Assumption 5.15) is the analytic condition that lets the contraction close.

What would settle it

Fix the model geometry $\mathbb{R} \times \hat Z_\zeta$ with the connection $\hat A + t\hat a_f$, where $\hat a_f = \frac{f}{\sqrt{7}}(\mathrm{id}\otimes_{\mathbb{R}}i)\,ds$, and search for $\mathbb{Z}_2$-invariant solutions of $L_{\hat A + t\hat a_f}L^*_{\hat A + t\hat a_f} b = 0$ with $\|b\|_{C^0_\beta} < \infty$ for some $\beta < 0$, or check directly whether the constant in Proposition 5.11 stays bounded as $f$ ranges over a compact interval $F \subset \mathbb{R}\setminus\pi\mathbb{Z}$. Finding a nonzero solution of that decay, or an estimate constant that blows up with $f$, would contradict Assumption 5.15 and collapse the family construction; finding none, as the paper expects, supports it.

Watch

Extended reading notes

Core claim

The paper's central claim, stated as Theorem A and Theorem B, is that families of $G_2$-instantons can be glued together and then deformed into genuine instantons even when the linearised instanton operator has a non-trivial cokernel. Theorem A is an abstract existence statement: for a degenerating family of $G_2$-manifolds $(\hat Y_t, \phi_t)$ arising from the generalised Kummer construction that resolves an orbifold with codimension-6 singular strata, and a finite group $H$ acting by coassociative-form-preserving isometries, any $H$-equivariant family of gluing data whose $H$-invariant cokernel of $L_{\tilde A} L^*_{\tilde A}$ vanishes deforms to a smooth $H$-invariant family of $G_2$-instantons with the same regularity as the starting family. Theorem B realises this for Joyce's Example 18 orbifold with $H = \mathbb{Z}_2$: for every compact interval $F \subset \mathbb{R} \setminus \pi\mathbb{Z}$ and all sufficiently small $t$, there is a compact $G_2$-manifold $(\hat Y, \phi_t)$ and an $\operatorname{SO}(14)$-bundle such that $f \mapsto A_{t,f}$ is a smooth family of $G_2$-instantons, each infinitesimally irreducible and non-flat, and the map $f \mapsto [A_{t,f}]$ into the moduli space of connections modulo gauge is injective. The mechanism that makes this possible is the $\mathbb{Z}_2$-action $y \mapsto -y$ on the orbifold: the one-dimensional obstruction space of the family is anti-invariant under it, so the fixed-point perturbation runs inside the $\mathbb{Z}_2$-invariant connections, a situation the paper describes as instantons being `$\mathbb{Z}_2$-unobstructed'.

Load-bearing premise

The construction rests on a linear estimate for the operator $L_{\tilde A} L^*_{\tilde A}$ that must hold with a constant independent of the family parameter $f$ and of the scale $t$; the proof of the underlying Schauder estimate (Proposition 5.11) is only sketched, and if the constant grows with $f$, or a $\mathbb{Z}_2$-invariant kernel appears, the fixed-point argument that builds the instanton family fails.

Editorial extensions

If this is right

  • Instantons on compact $G_2$-manifolds can form smooth families: the moduli space can contain curves, not only isolated points, even when each member is infinitesimally irreducible.
  • The gluing method of [Wal13a] now covers codimension-6 Kummer resolutions and families of connections with non-trivial cokernel: the obstruction to deforming a family can be cancelled by a finite group symmetry rather than by choosing the gluing data more carefully.
  • The constructed family injects into the moduli space modulo gauge, so it is a genuine curve of distinct gauge-equivalence classes of non-flat instantons, providing a concrete test object for any Donaldson–Thomas-type count of $G_2$-instantons.
  • Because the linearised operator has one-dimensional kernel (from the family direction) and the index over the odd-dimensional compact manifold is zero, each member of the family carries a one-dimensional obstruction space; the paper shows such obstructed instantons are nevertheless constructible when the obstructions are anti-invariant under a symmetry.

Reading between the lines

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

  • A general recipe suggests itself beyond the paper: whenever an orbifold-resolution admits a finite symmetry under which the obstruction space of a family of gluing data is anti-invariant, a smooth family of instantons should exist with parameter dimension bounded by the invariant kernel; testing this on other Joyce orbifolds or on Spin(7)-instantons would show how much of the mechanism is special
  • The paper's own caveat in Remark 6.21 leaves the resolved metric's identity open: the Calabi–Yau metric used here may not be isometric to the one in Joyce's original example, so the theorem may produce instantons on a $G_2$-manifold with a different $G_2$-structure from Joyce's, and whether the family persists on the exact Joyce metric remains an open question.
  • The injectivity proof yields a reusable criterion: a gauge equivalence between nearby members of a gluing family forces the derivative of the flat family to be an exact form in the limit $t \to 0$, and checking that derivative is a finite-dimensional representation-theoretic computation that could be repeated for other groups and other representations.
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

2 major / 3 minor

Summary. The paper develops an equivariant gluing construction for G2-instantons over generalized Kummer resolutions and applies it to Joyce's orbifold from [Joy96b, Example 18]. The main abstract result (Theorem A, comprising Theorem 3.13 and Propositions 4.5, 4.7, 5.16) gives conditions under which an H-equivariant family of almost-instantons on a Kummer resolution can be perturbed to a genuine family of G2-instantons, including estimates on derivatives with respect to the family parameter. The concrete result (Theorem B, comprising Propositions 6.31, 6.33, 6.34) produces, for sufficiently small scale t, an SO(14)-bundle over a compact G2-manifold arising from the orbifold and a smooth family of G2-instantons parametrized by a compact interval F in R minus pi Z, all infinitesimally irreducible and non-flat, with injective map into the moduli space. The proof combines Joyce's Kummer theorem, rigid Hermitian-Yang-Mills connections on ALE crepant resolutions due to Degeratu-Walpuski, a family fixed-point argument in weighted Holder spaces, and a detailed linear analysis using a Z2 symmetry to kill the obstruction space.

Significance. If established, this is the first smooth one-parameter family of G2-instantons over a compact G2-manifold, and it extends the existing gluing technology to codimension-six singular strata and to families whose linearized instanton operator has nontrivial cokernel. The paper is carefully structured and contains detailed proofs of most analytic steps, including the pregluing construction, the quadratic estimates, the family fixed-point argument, and the moduli-space injectivity argument; the explicit construction of the flat families in Section 6 and the nontrivial non-tangency result in Proposition 6.18 are valuable contributions. The paper is also transparent about caveats, notably the possible non-isometry of the resolution metric with Joyce's original example in Remark 6.21. These strengths make the overall approach credible. However, as stated, the central claims rest on two points that need further work: the uniform-in-f Schauder estimate is only sketched, and the injectivity statement needs a length restriction on the parameter interval or a quotient by the 2pi-periodicity.

major comments (2)
  1. [Section 5.1, Proposition 5.11] Proposition 5.11, which supplies the f-independent Schauder estimate for the model operator L L^* on R times the inverse image of the ball of radius t^{-1} kappa, is presented only as a proof sketch. The claimed t-independence and f-independence of the constant c_SI are asserted rather than proved: in the exterior region the estimate is reduced to equation (5.2) for the operator built from A_infty + tR(τ)^*a, where tR is only uniformly bounded and not small, and the lower-order terms tR(τ)^*a are said to be uniformly bounded without a displayed weighted-norm argument. This estimate is used in Proposition 5.17 and then in Proposition 5.16, which supplies assumption (2) of Theorem 3.13; the f-independence of the contraction constant in the fixed-point argument and the derivative estimates in Proposition 3.18 all inherit this uniformity. Since this is a load-bearing analytic input, the proof should be completed, or a precise reference covering this non-asymptotically-dilation-invariant family with a uniform constant should be supplied.
  2. [Section 6.5, Propositions 6.34 and 6.35; Remark 6.30] The injectivity statement in Proposition 6.34 is false for arbitrary compact intervals F in R minus pi Z of length at least 2pi. Remark 6.30 states that the preglued connections for parameters f_1 and f_2 are gauge equivalent whenever f_1 - f_2 is an integer multiple of 2pi, and by the uniqueness in the fixed-point construction the same periodicity passes to the perturbed instantons. The proof of Proposition 6.35(1) rules out distinct limiting theta-values via Proposition 6.3, but it does not exclude f'_infty = f_infty + 2pi, which gives the same value of theta in S^1. Consequently Theorem B as stated needs the additional hypothesis that F has length strictly less than 2pi, or the injectivity claim must be formulated for the induced curve on R/2piZ. This is a load-bearing correction to the main example.
minor comments (3)
  1. [Section 5.2, Proposition 5.16, and Section 6.4, Proposition 6.29] Proposition 5.16 and the corresponding bullet in Proposition 6.29 state the linear estimate for b in Omega^1 direct sum Omega^7, but the operator L L^* acts on Omega^6 direct sum Omega^0, as used in Theorem 3.13(2) and in Propositions 5.17 and 5.18. The domain in these statements should be corrected.
  2. [Remark 6.21] The caveat that the Calabi-Yau metric used for the resolution may not be isometric to the one in Joyce's original example is explicitly acknowledged, but Theorem B should state in its formulation that the G2-manifold is one arising from the orbifold via the Kummer construction and is not claimed to be isometric to Joyce's original resolved example.
  3. [Section 6.2, Proposition 6.13 and Definition 6.16] The passage from the family of flat bundles to the fixed bundle E_0 uses isomorphisms F'_f and F_f; a brief sentence explaining why the modified F_f is still orientation preserving and well-defined on the quotient after the cut-off modification would improve readability.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the construction is a genuine gluing argument with independent inputs; the only concerns are a sketched analytic estimate and an acknowledged metric caveat, neither of which makes the derivation circular.

full rationale

The central claim (Theorem B via Theorem A) is a gluing construction whose inputs are external and independent: Joyce's generalised Kummer theorem (Theorem 2.8), Walpuski's analytic gluing framework (Theorem 3.13 and the estimates in Propositions 4.5 and 4.7), and Degeratu-Walpuski's rigid Hermitian Yang-Mills connections (Proposition 2.21 and its SO(2n)-extension in Proposition 2.25). The family parameter f is a genuine geometric parameter: f parametrises flat connections A0 + a0,f on a fixed SO(14)-bundle, and the proof of injectivity of f into the moduli space relies on Proposition 6.18, which shows that the derivative of the family is not tangent to the gauge orbit. That statement is proved from the explicit representation fθ and the fixed-point structure, so injectivity is not built into the definition of the family. The author's self-citations, chiefly [Gut24a] and [Gut24b], are used for background conventions, notation, and elementary facts, and the main theorem does not reduce to them. The two issues that a careful reader should flag are (1) Proposition 5.11 is only a proof sketch: the f-independence of the Schauder constant for L_{\hat A+t\hat a_f}L^* is asserted rather than fully displayed, and this estimate underpins the linear estimate in Proposition 5.16 and hence the contraction argument in Theorem 3.13; this is a genuine completeness or correctness gap, but it is not circularity, because the estimate is a quantitative analytic input about a model operator and does not presuppose the existence or properties of the instantons being constructed. (2) Remark 6.21 explicitly acknowledges that the Calabi-Yau metric used for the resolution may not be isometric to Joyce's original resolved metric; this is an honest limitation about the relation to Joyce's example, not a circular step, and Theorem B is stated accordingly as producing a G2-manifold arising from the orbifold rather than claiming isometry with Joyce's metric. No equation or proposition in the paper reduces by construction to its own input, and no load-bearing argument is smuggled in through self-citation. Therefore the paper receives a circularity score of 0.

Assumptions & free parameters 1 free parameters · 5 assumptions · 0 invented entities

The central claim relies on external theorems for the existence of G2-structures and rigid HYM connections, plus in-paper representation-theoretic computations. No new physical entities are introduced. The most fragile input is the sketched Schauder estimate in Proposition 5.11.

free parameters (1)
  • Generic stability parameter ζ
    The paper fixes one generic ζ in the GIT-quotient description of the crepant resolution (Section 6.3). The results are stated for any such generic choice, so this is a construction choice rather than a fitted value; it does not enter the estimates numerically.
assumptions (5)
  • domain assumption Joyce's generalized Kummer theorem (Theorem 2.8) provides, for small t, a torsion-free G2-structure ϕ_t on the resolution with ||ϕ_t - \tilde ϕ_t||_{C^{k,α}_{0,t}} < c t^{1/2}, and for codimension-6 strata this improves to t^{5/2} (Remark 2.9).
    Invoked in Sections 3 and 4 as the geometric input for all analytic estimates; the paper cites [Joy96a, Joy96b, Joy00, Wal13a] rather than reproving it.
  • domain assumption There exist infinitesimally rigid, asymptotically flat HYM connections on ALE crepant resolutions C^3/Γ (Proposition 2.21, from [DW16]), and these remain rigid after realification to SO(2n) bundles (Proposition 2.25).
    Used in Sections 5 and 6 to satisfy Assumption 5.15; the paper cites [DW16] and proves Proposition 2.25 in text.
  • standard math The flat representation f_θ of Γ into SO(14) constructed in Proposition 6.2 exists and has the stated fixed-point properties (Lemma 6.5, Propositions 6.6, 6.10).
    Verified by direct computation in Section 6.1; these properties underpin the Z2-equivariance and the vanishing of the H-invariant cokernel.
  • ad hoc to paper The family of flat connections (A0 + a0,f)_f is not tangent to the gauge orbit at any f (Proposition 6.18).
    Needed in Proposition 6.34 to prove injectivity of the curve in moduli space; proven in the paper by an explicit L2-orthogonality argument.
  • domain assumption The t-uniform Schauder estimate for the family-dependent model operator (Proposition 5.11) holds with an f-independent constant.
    The proof is only sketched; this is the most fragile analytic input and is flagged as a red flag.

how reviews work

0 comments
Cite this review

Pith. "Pith review of A smooth family of $G_2$-instantons over a generalised Kummer construction." pith.science (2026). https://pith.science/paper/NIJSQGU5

@misc{pith2026250700655,
  author       = {Pith},
  title        = {Pith review of: A smooth family of $G_2$-instantons over a generalised Kummer construction},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/NIJSQGU5}},
  note         = {Machine review of arXiv:2507.00655}
}
abstract

We construct a smooth 1-parameter family of $G_2$-instantons over a generalised Kummer construction desingularising a $G_2$-orbifold discovered by Joyce. For this we extend the gluing construction for $G_2$-instantons developed by Walpuski to Kummer constructions resolving $G_2$-orbifolds whose singular strata are of codimension 6 and to connections (and entire families of connections) whose linearised instanton operator has a non-trivial cokernel. In order to overcome the corresponding obstructions, we utilise a $\mathbb{Z}_2$-action on the ambient manifold. More precisely, we perturb the (family of) pre-glued almost-instantons inside the class of $\mathbb{Z}_2$-invariant connections, which has the advantage that only the $\mathbb{Z}_2$-invariant locus of the cokernel needs to vanish. We then prove that the instantons that we construct over the resolution of the orbifold found by Joyce are all infinitesimally rigid and non-flat. Moreover, we show that the resulting curve into the moduli space of $G_2$-instantons modulo gauge is injective, that is, no two distinct instantons within the family are gauge-equivalent. To the author's knowledge this is the first example of a smooth 1-parameter family of instantons over a compact $G_2$-manifold.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

40 extracted references · 39 canonical work pages

  1. [1]

    Orbifolds and stringy topology , volume 171 of Camb

    Alejandro Adem, Johann Leida, and Yongbin Ruan. Orbifolds and stringy topology , volume 171 of Camb. Tracts Math. Cambridge: Cambridge University Press, 2007

    work page 2007

  2. [2]

    The mass of an asymptotically flat manifold

    Robert Bartnik. The mass of an asymptotically flat manifold. Commun. Pure Appl. Math. , 39:661--693, 1986

    work page 1986

  3. [3]

    Deformations and desingularizations of conically singular associative submanifolds

    Gorapada Bera. Deformations and desingularizations of conically singular associative submanifolds. preprint, arXiv :2310.01741 , 2023. https://arxiv.org/abs/2310.01741

    arXiv 2023

  4. [4]

    The McKay correspondence as an equivalence of derived categories

    Tom Bridgeland, Alastair King, and Miles Reid. The McKay correspondence as an equivalence of derived categories. J. Am. Math. Soc. , 14(3):535--554, 2001

    work page 2001

  5. [5]

    Corrigan, C

    E. Corrigan, C. Devchand, D.B. Fairlie, and J. Nuyts. First-order equations for gauge fields in spaces of dimension greater than four. Nuclear Physics B , 214(3):452--464, 1983

    work page 1983

  6. [6]

    Flops of \(G\) - Hilb and equivalences of derived categories by variation of GIT quotient

    Alastair Craw and Akira Ishii. Flops of \(G\) - Hilb and equivalences of derived categories by variation of GIT quotient. Duke Math. J. , 124(2):259--307, 2004

    work page 2004

  7. [7]

    S. K. Donaldson and P. B. Kronheimer. The geometry of four-manifolds . Oxford Math. Monogr. Oxford: Clarendon Press, 1990

    work page 1990

  8. [8]

    Associative submanifolds in Joyce 's generalised Kummer constructions

    Shubham Dwivedi, Daniel Platt, and Thomas Walpuski. Associative submanifolds in Joyce 's generalised Kummer constructions. Commun. Math. Phys. , 401(3):2327--2353, 2023

    work page 2023

Show all 40 references
  1. [9]

    Gauge theory in higher dimensions

    Simon Donaldson and Ed Segal. Gauge theory in higher dimensions. II . In Geometry of special holonomy and related topics , page 1–41. Somerville, MA: International Press, 2011

    2011

  2. [10]

    S. K. Donaldson and R. P. Thomas. Gauge theory in higher dimensions. In The geometric universe: science, geometry, and the work of Roger Penrose. Proceedings of the symposium on geometric issues in the foundations of science, Oxford, UK, June 1996 in honour of Roger Penrose in...

    1996

  3. [11]

    Rigid HYM connections on tautological bundles over ALE crepant resolutions in dimension three

    Anda Degeratu and Thomas Walpuski. Rigid HYM connections on tautological bundles over ALE crepant resolutions in dimension three. SIGMA, Symmetry Integrability Geom. Methods Appl. , 12:paper 017, 23, 2016

    2016

  4. [12]

    On counting associative submanifolds and Seiberg - Witten monopoles

    Aleksander Doan and Thomas Walpuski. On counting associative submanifolds and Seiberg - Witten monopoles. Pure Appl. Math. Q. , 15(4):1047--1133, 2019

    2019

  5. [13]

    S \'a Earp and Thomas Walpuski

    Henrique N. S \'a Earp and Thomas Walpuski. \(G_ 2 \) -instantons over twisted connected sums. Geom. Topol. , 19(3):1263--1285, 2015

    2015

  6. [14]

    Freed and Karen K

    Daniel S. Freed and Karen K. Uhlenbeck. Instantons and four-manifolds , volume 1 of Math. Sci. Res. Inst. Publ. Cambridge University Press, Cambridge; Mathematical Sciences Research Institute, Berkeley, CA, 1984

    1984

  7. [15]

    Spin(7) -instantons on Joyce's first examples of compact Spin(7) -manifolds

    Mateo Galdeano, Daniel Platt, Yuuji Tanaka, and Luya Wang. Spin(7) -instantons on Joyce's first examples of compact Spin(7) -manifolds. preprint, arXiv :2310.03451 , 2023. https://arxiv.org/abs/2310.03451

    2023

  8. [16]

    Coassociative submanifolds and G2-instantons in Joyce’s generalised Kummer constructions

    Dominik Gutwein. Coassociative submanifolds and G2-instantons in Joyce’s generalised Kummer constructions. dissertation , 2024. https://edoc.hu-berlin.de/items/26a7b79d-5f84-43a7-973d-1dd9b0e4b608

    2024

  9. [17]

    Coassociative submanifolds in Joyce 's generalised Kummer constructions

    Dominik Gutwein. Coassociative submanifolds in Joyce 's generalised Kummer constructions. Pure Appl. Math. Q. , 20(2):923--954, 2024

    2024

  10. [18]

    \(G_2\) instantons and the Seiberg - Witten monopoles

    Andriy Haydys. \(G_2\) instantons and the Seiberg - Witten monopoles. In Gromov-Witten theory, gauge theory and dualities, Kioloa, Australia, January 6--16, 2016 , page 12. Canbarra: The Australian National University, Centre for Mathematics and its Applications, 2019. Id/No 3

    2016

  11. [19]

    Complex geometry

    Daniel Huybrechts. Complex geometry. An introduction . Universitext. Berlin: Springer, 2005

    2005

  12. [20]

    Partial Resolutions of Orbifold Singularities via Moduli Spaces of HYM -type Bundles

    Alexander V Sardo Infirri. Partial Resolutions of Orbifold Singularities via Moduli Spaces of HYM -type Bundles . preprint, arXiv :9610004 , 1996. https://arxiv.org/abs/alg-geom/9610004

    1996 arXiv

  13. [21]

    Joyce and Spiro Karigiannis

    Dominic D. Joyce and Spiro Karigiannis. A new construction of compact torsion-free \(G_ 2 \) -manifolds by gluing families of Eguchi - Hanson spaces. J. Differ. Geom. , 117(2):255–343, 2021

    2021

  14. [22]

    Dominic D. Joyce. Compact Riemannian 7-manifolds with holonomy \(G_ 2\) . I . J. Differ. Geom. , 43(2):291--328, 1996

    1996

  15. [23]

    Dominic D. Joyce. Compact Riemannian 7-manifolds with holonomy \(G_ 2\) . II . J. Differ. Geom. , 43(2):329--375, 1996

    1996

  16. [24]

    Dominic D. Joyce. Compact manifolds with special holonomy . Oxford Math. Monogr. Oxford: Oxford University Press, 2000

    2000

  17. [25]

    Kronheimer and Hiraku Nakajima

    Peter B. Kronheimer and Hiraku Nakajima. Yang- Mills instantons on ALE gravitational instantons. Math. Ann. , 288(2):263--307, 1990

    1990

  18. [26]

    Twisted connected sums and special Riemannian holonomy

    Alexei Kovalev. Twisted connected sums and special Riemannian holonomy. J. Reine Angew. Math. , 565:125--160, 2003

    2003

  19. [27]

    Lockhart and Robert C

    Robert B. Lockhart and Robert C. McOwen. Elliptic differential operators on noncompact manifolds. Ann. Sc. Norm. Super. Pisa, Cl. Sci., IV. Ser. , 12:409--447, 1985

    1985

  20. [28]

    D. G. Markushevich, M. A. Olshanetskij, and A. M. Perelomov. Description of a class of superstring compactifications related to semi- simple Lie algebras. Commun. Math. Phys. , 111:247--274, 1987

    1987

  21. [29]

    Moduli spaces of anti-self-dual connections on ALE gravitational instantons

    Hiraku Nakajima. Moduli spaces of anti-self-dual connections on ALE gravitational instantons. Invent. Math. , 102(2):267--303, 1990

    1990

  22. [30]

    Louis Nirenberg and Homer F. Walker. The null spaces of elliptic partial differential operators in R \(^n\) . J. Math. Anal. Appl. , 42:271--301, 1973

    1973

  23. [31]

    Goncalo M. F. Oliveira. Monopoles in higher dimensions. dissertation , 2014. https://spiral.imperial.ac.uk/handle/10044/1/23570

    2014

  24. [32]

    \(G_2\) -instantons on resolutions of \(G_2\) -orbifolds

    Daniel Platt. \(G_2\) -instantons on resolutions of \(G_2\) -orbifolds. Commun. Math. Phys. , 405(3):63, 2024. Id/No 81

    2024

  25. [33]

    okova geometry-topology conference, G\

    Dietmar A. Salamon and Thomas Walpuski. Notes on the octonions. In Proceedings of the 23rd G\"okova geometry-topology conference, G\"okova, Turkey, May 30 -- June 3, 2016 , pages 1--85. Somerville, MA: International Press; G \"o kova: G \"o kova Geometry-Topology Conferences (...

    2016

  26. [34]

    Gauge theory and calibrated geometry

    Gang Tian. Gauge theory and calibrated geometry. I . Ann. Math. (2) , 151(1):193–268, 2000

    2000

  27. [35]

    Complete K \"a hler manifolds with zero Ricci curvature

    Gang Tian and Shing Tung Yau. Complete K \"a hler manifolds with zero Ricci curvature. II . Invent. Math. , 106(1):27--60, 1991

    1991

  28. [36]

    \(G_ 2 \) -instantons on generalised Kummer constructions

    Thomas Walpuski. \(G_ 2 \) -instantons on generalised Kummer constructions. Geom. Topol. , 17(4):2345–2388, 2013

    2013

  29. [37]

    Gauge theory on G2–manifolds

    Thomas Walpuski. Gauge theory on G2–manifolds . dissertation , 2013. http://hdl.handle.net/10044/1/14365

    2013

  30. [38]

    \(G_2\) -instantons, associative submanifolds and Fueter sections

    Thomas Walpuski. \(G_2\) -instantons, associative submanifolds and Fueter sections. Commun. Anal. Geom. , 25(4):847--893, 2017

    2017

  31. [39]

    Moduli of \(G\) -constellations and crepant resolutions

    Ryo Yamagishi. Moduli of \(G\) -constellations and crepant resolutions. II : The Craw - Ishii conjecture. Duke Math. J. , 174(2):229--285, 2025

    2025

  32. [40]

    Yau and Yung Yu

    Stephen S.-T. Yau and Yung Yu. Gorenstein quotient singularities in dimension three , volume 505 of Mem. Am. Math. Soc. Providence, RI: American Mathematical Society (AMS), 1993

    1993

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.