Diffeological non-Abelian Hodge theory: relative harmonic metrics and deformation theory

Mahmud Azam; Steven Rayan

REVIEW 1 major objections 4 minor 16 references

Diffeological non-Abelian Hodge theory: relative harmonic metrics and deformation theory

T0 review · 1 major / 4 minor · reviewed 2026-08-01 · deepseek-v4-flash

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

Pith's one-line read This paper proves that for every smooth family of stable Higgs bundles on a compact Kähler manifold satisfying the standard numerical conditions, a harmonic metric can be chosen smoothly in the family parameter, making the non-Abelian Hodge

desk verdict Solid analytic core for stable families, with genuinely new plotwise smoothness and variation formulas; the negative polystable examples are the part to scrutinize. read the letter →

arxiv 2607.18989 v1 pith:JRCGFGUZ submitted 2026-07-21 math.DG math.AGmath.APmath.CTmath.RT

Mahmud Azam , Steven Rayan This is my paper · ORCID

classification math.DGmath.AGmath.APmath.CTmath.RT

keywords diffeologicalstacksharmonicmetricsHiggsbundlesnon-AbelianHodgecorrespondencerelativedeformationtheorySobolevcompletionspolystablefamiliesHitchin–Simpsonequation

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 that the fibrewise harmonic metric of a stable Higgs bundle varies smoothly in the parameter of a smooth family, provided the usual non-Abelian Hodge numerical conditions hold. The proof runs the Hitchin–Simpson moment-map equation on Sobolev completions over arbitrary finite-dimensional parameter plots, uses stability to make the linearized Jacobi operator invertible on a trace-free slice, and glues local solutions via a determinant normalization. From this existence-plus-regularity result the paper derives that the non-Abelian Hodge transform is smooth along diffeological plots and computes its first variation explicitly via a Green operator. On the polystable locus the paper shows the analogous statement fails: real-analytic families can lack continuous harmonic metrics, so the correspondence is preserved only through a larger weak C^0 operator-level mediator. For semistable families, the extension-generated stack is characterized by relative harmonic filtrations, with an obstruction theory for assembling them.

What carries the argument

The load-bearing machinery is the parameter-dependent Banach-space formulation of the Hitchin–Simpson moment map. A family of Higgs bundles is viewed as a smooth map into an ambient Sobolev affine space; the harmonic equation becomes a nonlinear map F(u,s) on fixed Sobolev spaces, where s is the logarithmic metric variation in an exponential gauge. The metric-direction derivative is the Jacobi operator L_h = (D'')*_h D'' (the Higgs Laplacian on Hermitian endomorphisms), which is self-adjoint, elliptic, and nonnegative. On a stable Higgs bundle its kernel consists of scalar Hermitian Higgs endomorphisms, so after imposing a determinant normalization and projecting to the trace-free slice, L_h

What would settle it

Search for a smooth family of stable Higgs bundles over a compact parameter manifold, satisfying the numerical conditions, whose normalized fibrewise harmonic metric is not smooth in the parameter. Theorem 1.1 asserts that no such family exists; a counterexample (constructed, say, by inducing a harmonic metric through a family of gauge transformations whose smoothness degenerates) would refute the central claim. Alternatively, for a concrete family with all assumptions satisfied, compute the curvature tensor of the associated connection and verify the Chern–Weil identity; if the curvature ener

Watch

Extended reading notes

Core claim

The central discovery is that the non-Abelian Hodge correspondence, which classically relates stable Higgs bundles to irreducible flat connections fibrewise, is itself a smooth operation on families: given a smooth family (E, D'') of stable Higgs bundles over a compact Kähler manifold X parametrized by a smooth manifold U, with all fibres satisfying the numerical conditions ν1=ν2=0, there exists a global smooth Hermitian metric h on E such that h_u is harmonic for each fibre. After fixing a Hermitian–Einstein determinant metric, the harmonic metric is unique, so local solutions glue globally. The proof converts the Hitchin–Simpson equation into a nonlinear map on Sobolev spaces whose metric

Load-bearing premise

The theorem's flatness conclusion depends on the exact numerical conditions ν1(E)=0 and ν2(E)=0 holding for every fibre; the analytic construction only solves the trace-free moment-map equation, and without those Chern-Weil identities the resulting Hermitian–Einstein–Higgs metric need not be flat, so 'harmonic' in the paper's strong sense would fail even though the smooth family of metrics exists.

Editorial extensions

If this is right

The non-Abelian Hodge transform is a smooth morphism of diffeological stacks on the stable locus, with smooth inverse on irreducible families, so the classical correspondence holds at the level of smooth families, not just pointwise.
The first variation formula gives a concrete Green-operator expression for the infinitesimal transform along any plot, agreeing with the classical cohomological comparison at unobstructed points.
Stable loci lie in the harmonic-image substack before any extension completion or stackification, answering the stable case of the earlier open question about smooth parameter dependence.
The smooth Hodge λ-family on the stable locus supplies a finite-dimensional-parameter family of λ-connections, with pullback functoriality, giving a diffeological Hodge enhancement.
The paper's negative examples for polystable families and the weak C^0 mediator show that extension completion is genuinely necessary outside the stable/constant-type locus, and that the correspondence can persist in a weaker operator-level sense even when continuous harmonic reductions fail.

Reading between the lines

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

The analytic machinery is likely robust enough to handle families with additional structure (e.g., parabolic or twisted Higgs bundles) where the same Banach-space implicit-function argument would yield analogous smooth dependence, though the flatness upgrade would need the appropriate numerical conditions.
The spectral-gap perspective in the paper suggests a quantitative version: the norm of the first variation of the harmonic metric near a polystable degeneration is controlled by the inverse spectral gap, indicating that smoothness degrades in a predictable way as the stabilizer grows; this could be made into a testable regularity estimate.
The weak C^0 operator-level mediator, although not a smooth transform, might be a natural object for studying families of semisimple local systems that cross separatrices or non-closed orbits, where the usual harmonic metric degenerates but the adjoint operators still converge.
If the numerical conditions are relaxed, the paper's proof still yields smooth solutions to the moment-map equation (not necessarily flat); these could be interpreted as 'almost harmonic' metrics and might be used to construct approximate Hodge systems with controlled error, quantifying the role of the Chern-class conditions.

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.

Desk Editor's Note

read the letter

Colleague,

You should know this is a real sequel with a substantial analytic theorem, and the arguments I could audit hold together. The headline result is a global smooth harmonic metric for any smooth stable Higgs family over an arbitrary finite-dimensional plot, with the first-variation Green-operator formulas and a smooth diffeological transform on the stable locus. That is genuinely new in its plotwise form, and the paper is honest that the local analytic dependence was already known from Hu–Sun–Yang–Zuo and Kim–Wilkin; the novelty is the globalization, the stack-level formulation, and the deformation theory.

The analytic core is standard but carefully done. The Jacobi operator identity, stability-to-simplicity, the fixed-Banach-space IFT, and the gluing are all consistent as far as I could audit. The constant-type polystable theorem is a nice clean statement.

One worry circulated in our stress test was that the flatness upgrade from a moment-map solution needs a Chern-Weil identity that isn't written. I don't think this worry lands. They define ν2 via ch2, and the standard Higgs curvature energy identity says the L2 norm of the primitive curvature is proportional to the integral of ch2 wedge ω^{n-2}. With μ=0, that's exactly the identity; no extra Hodge-index step is required. The paper's shorthand "standard Chern-Weil identity" is terse, but the logic is sound. A referee should still ask them to write the identity explicitly, because it's the only place the numerical conditions enter.

The real soft spot is the negative material in §14.3–14.4. The abstract and introduction make strong claims: real-analytic polystable families with no continuous harmonic metric, no relative harmonic filtration, and the weak C0 mediator with a stack equivalence. Those are the headline news for the polystable side, but they're in the final section, and I couldn't verify them in detail. The rest of the paper doesn't depend on them, so the stable and constant-type results stand regardless. If the examples are right, the paper is a significant contribution to the family-level non-abelian Hodge story; if they're wrong, the polystable claims collapse but the stable part survives.

The main dependency is on the prior paper's diffeological stack framework. If you don't buy that framework, the stack-level consequences won't matter much to you, but the analytic theorems are independent.

Verdict: This deserves a serious referee. It's long and could be tightened, and the examples need careful checking, but the core is original and sound. I'd send it to review.

Referee Report

1 major / 4 minor

Summary. This paper develops the analytic foundations of the authors' earlier diffeological approach to non-Abelian Hodge theory. It proves that smooth families of stable Higgs bundles over arbitrary finite-dimensional parameter manifolds admit globally smooth harmonic metrics, provided each fibre satisfies the numerical conditions (9). The proof proceeds by local trivialization, Sobolev completions, a Banach implicit-function theorem for the trace-free moment-map equation, determinant normalization, and gluing by uniqueness. It then derives the first-variation formula L_h s_0 = -S_h(eta), identifies the differential of the transform, treats locally split polystable families of constant type, introduces relative harmonic filtrations with an obstruction theory, extends the results to finite regularity and reduced singular parameter spaces, discusses heat-flow limits, and constructs a smooth Hodge lambda-family on the stable locus. The negative examples in Section 14.3 are used to define a weak C0 operator-level harmonic mediator.

Significance. If the analytic core is correct, the paper fills a substantial gap in the diffeological framework: it proves that stable families lie in the essential image of the harmonic mediator before extension completion, and it gives a plotwise smooth non-Abelian Hodge transform. The fixed-Banach-space formulation (eqs. 24-25), the identification of the Jacobi operator as (D'')*D'' (Prop. 4.2), the stability-to-simplicity argument (Prop. 4.4), the determinant Poisson step (Prop. 5.1), the normalized uniqueness via Donaldson-functional convexity (Prop. 5.5), and the gluing argument (Cor. 5.6) are clearly presented and auditable. The paper is honest in separating the moment-map problem from the flatness upgrade, and the explicit examples of polystable families without continuous harmonic metrics, if correct, are a strong and falsifiable contribution. The first-variation and Green-operator formulas provide useful tools for deformation theory.

major comments (1)

[§5.2–5.3, Remark 5.3, eqs. (8)–(9)] The flatness upgrade is not proved. The text invokes a 'standard Hitchin–Simpson Chern–Weil identity' but never writes it. The relevant characteristic-number combination is the Bogomolov-type expression ∫(2r c2−(r−1)c1^2)∧ω^{n−2}, not ∫ch2∧ω^{n−2} alone. As written, (9) sets only ν1,ν2 to zero; the paper does not supply the Hodge-index/Bogomolov argument that this forces the curvature energy to vanish. Since Theorem 1.1's harmonic conclusion and all subsequent plotwise/stack results depend on this step, the proof is incomplete. Please state the identity in the paper's conventions and either prove the implication from (9) for stable fibres or strengthen (9) to full Chern-class vanishing.

minor comments (4)

[§3.1, eq. (24)] The definition of the trace-free target T^0_k depends on the fixed background h0, while the source uses the moving metric. This is clarified by the isometry I_s, but a one-sentence reminder before (24) would help readers.
[§5.3, proof of Theorem 1.1] The phrase 'the standard Chern–Weil identity expresses the remaining curvature energy ... as the characteristic-number combination in (9)' is imprecise: the identity involves a linear combination of c1 and c2 (or ch2), not the two separate numbers ν1 and ν2. This is the same point as the major comment; please rewrite this sentence after adding the explicit identity.
[§11.1, Definition 11.1] The definition of mixed regularity would be clearer if phrased as: in local trivializations, the coefficient map u ↦ coefficient(u,·) is C^d into the Fréchet space C^∞(X), rather than saying the coefficients themselves are 'smooth in X and C^d in u'.
[§7.4, Proposition 7.11] For the endpoint j=0, the use of W^{-1,2} and elliptic duality is invoked without a precise definition or reference. Adding a short explanation or a citation for the negative-order Sobolev spaces would improve readability.

Circularity Check

0 steps flagged · score 2.0 of 10

No significant circularity: the core analytic derivation is self-contained; self-citations to [2] supply context and stack language but are not load-bearing for Theorem 1.1 or the first-variation formulas.

full rationale

The paper's main derivation chain is genuinely self-contained. Theorem 1.1 is proved by a local Banach implicit-function argument on the trace-free Hitchin–Simpson moment map, a scalar Poisson construction for the determinant normalization, parameter-dependent elliptic bootstrap, and normalized gluing via a convexity uniqueness argument. None of these steps assumes the conclusion; each is established in the paper against classical external benchmarks (elliptic regularity, scalar Poisson theory, the pointwise Hitchin–Simpson correspondence). The first-variation identities L_h s = -S_h(eta) and s = -G_h S_h(eta) are obtained by differentiating the solved moment-map equation, not by restating an input. The numerical conditions (9) are hypotheses under which the classical Chern--Weil identity promotes a moment-map solution to flatness; this step is explicitly separated in Remark 5.3 and depends on the standard compact-Kähler mechanism, not on the target family statement. The relative harmonic filtration criterion (Theorem 1.6 / 9.3) is a direct geometric reformulation of the extension-completion definition from [2] and is acknowledged as formally close to that construction; it is not used to generate the analytic existence results. Self-citations to [2] provide the diffeological stack framework, terminology, and open questions, but the stable existence theorem, regularity, uniqueness, polystable stratum results, and variation formulas do not reduce to assertions imported from [2]. The skeptical concern about the exact form of the Chern--Weil identity is a potential gap or correctness risk in the topological/flatness step, not a circularity: no fitted parameter is renamed as a prediction and no definition is constructed in terms of the result.

Assumptions & free parameters 2 free parameters · 8 assumptions · 3 invented entities

No empirical data are fitted: this is pure mathematics. The two listed 'free parameters' are technical choices (Sobolev index, convention factors) that any admissible value satisfies. The axioms separate (a) standard elliptic/Kähler machinery (items 1, 2, 7, 8), (b) external mathematical content consumed by the paper (pointwise NAH theorem; numerical conditions; reductive flat-bundle classifications; items 3–5), and (c) the paper's own methodological device (local trivialization into a fixed Sobolev affine space; item 6). The most load-bearing external inputs are the classical pointwise theorem (item 3) and the ν2 = 0 flatness upgrade (item 4, flagged in Remark 5.3); the most distinctive internal device is solving the moment-map equation on the ambient non-integrable coefficient space before restriction to integrable data.

free parameters (2)

Sobolev regularity index k = any integer with k > n+1 (eq. (21))
Chosen so that W^{k,2} is a Banach algebra and the nonlinear moment map is smooth (Lemma 3.1). The theorems are independent of the particular admissible value, so this is a technical device, not a tuned parameter.
Moment-map normalization convention factors (√-1 Λ and related constants) = fixed convention factors absorbed into definitions (eqs. (10), (14), (30))
Affect the explicit constant in the Chern–Weil identity and the Donaldson functional convexity (Prop. 5.5). All results are convention-independent; listed for exhaustiveness.

assumptions (8)

standard math Kähler identity and harmonic-bundle identities on a compact Kähler manifold: (D'')*_h = −√-1[Λ, D'_h], with flatness giving [D'', D'_h]_gr = 0 (Prop. 4.2, eq. (29)).
Converts the metric-direction derivative of the moment map into the self-adjoint Laplacian L_h = (D'')*D''; the kernel identity (Cor. 4.3) and all Green-operator arguments use it.
standard math Fredholm/elliptic theory on compact X: Sobolev multiplication for k > n+1; scalar Laplacian is an isomorphism on mean-zero functions (eq. (21), Prop. 5.1, eq. (44)).
Underpins the fixed-Banach-space implicit-function theorem (Lemma 3.1, Thm 5.2) and the global Hermitian–Einstein determinant metric construction.
domain assumption Classical pointwise non-Abelian Hodge theorem (Hitchin–Donaldson–Corlette–Simpson): stable fibres admit harmonic metrics; polystable corresponds to existence; irreducible flat corresponds to stable Higgs (used in Thm 1.1 proof, Thm 6.6, and slicewise in §14.4).
Every slice-level existence statement, the inverse transform, and the slicewise-harmonic condition of the weak C0 mediator invoke this external theorem.
domain assumption NAH numerical conditions ν1(E_u) = ν2(E_u) = 0 with ν2 via ch2 (eqs. (8)–(9)); the Chern–Weil identity upgrades moment-map solutions to flatness (Remark 5.3).
A moment-map solution alone is not flat without the exact numerical conditions; the paper's own Remark 5.3 isolates this dependency, which is the paper's weakest load-bearing premise.
domain assumption External classification facts: harmonic metrics on reductive flat bundles have parallel endomorphism algebra; irreducible flat implies stable (used in Thm 8.2 converse and Thm 6.6).
Used to prove that harmonic metrics on constant-type polystable families are exactly the product-form metrics (70) and to prove the inverse transform on the irreducible locus.
ad hoc to paper Local trivialization device: after choosing a parameter-direction connection, smooth Higgs families become smooth maps into a fixed Sobolev affine coefficient space (Remark 2.2, eqs. (18)–(19)).
This ambient-affine-space setup, used again for singular bases in §11.3, is the paper's own methodological premise rather than an established theorem about all families; it is needed to place the nonlinear equation on fixed Banach spaces.
standard math Uniform strong ellipticity with uniform estimates for the family of linearized operators (eq. (27), Prop. 3.2/Appendix A); invertibility persists under small perturbations via the Neumann series (eq. (28)).
Provides joint smoothness (Prop. 5.4) and the finite-regularity bootstrap (§11). The uniform spectral-gap hypothesis is exactly what fails approaching polystable degeneration (eq. (102), §5.4, Prop. 11.14).
standard math Stability of every fibre implies simplicity: every Higgs endomorphism of a stable Higgs bundle is scalar (Prop. 4.4, Remark 4.5), converting stability into invertibility of the trace-free Jacobi operator.
The single point where slope stability enters the Banach IFT. At a merely polystable fibre with nontrivial centralizer, ker L_h is nonzero (Prop. 4.9), and the paper's §14 examples show this failure is essential rather than technical.

invented entities (3)

Weak C0 operator-level harmonic mediator (prestack)
purpose: Carry the Dolbeault–de Rham correspondence for degenerate polystable families where continuous harmonic metrics and relative harmonic filtrations fail, by requiring the harmonic metric only slicewise while adjoint Higgs operators and flat connections remain continuous (Abstract; §1.3; §14.4).
The claim that it strictly enlarges the metric-regular mediator rests on internal examples (§14.4) not auditable in the provided text, and the resulting equivalence is partly built into the construction (both endpoint stacks are images of the same mediator prestack). It thus supplies no falsifiable handle independent of this paper's definitions.
Relative harmonic filtration independent evidence
purpose: Geometric characterization of membership in the extension-generated stacks M^H_Dol(X) and M^H_dR(X) (Def. 1.5, Thm 1.6/9.3); upgraded to C^d versions in Def. 11.5.
Its defining property — a filtration by smooth invariant subbundles with harmonic quotients — is an externally checkable condition on any family, and the paper supplies both positive (Thm 9.3, Cor. 9.5) and negative (Prop. 14.3) instances.
Parameter-singular (boundary) harmonic metrics
purpose: Described as the natural boundary objects when a stable family degenerates to a non-polystable semistable fibre (Prop. 11.13), and as bookkeeping in the finite-regularity analysis (§11.3, Remark 11.15).
The paper explicitly states it does not construct such boundary metrics ('It does not construct such a boundary metric'); they enter only as the contradiction target of Prop. 11.13 and as a category of expected phenomena.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Diffeological non-Abelian Hodge theory: relative harmonic metrics and deformation theory." pith.science (2026). https://pith.science/paper/JRCGFGUZ

@misc{pith2026260718989,
  author       = {Pith},
  title        = {Pith review of: Diffeological non-Abelian Hodge theory: relative harmonic metrics and deformation theory},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/JRCGFGUZ}},
  note         = {Machine review of arXiv:2607.18989}
}

abstract

Let $X$ be a compact K\"ahler manifold. In prior work, we constructed diffeological moduli stacks of Higgs and flat bundles on $X$, related by extension completion of smooth harmonic families. Here, we develop the relative analytic theory. On Sobolev completions over arbitrary plots, we prove that every smooth stable Higgs family satisfying the numerical conditions admits a global smooth harmonic metric. Fixing a Hermitian--Einstein determinant metric removes scalar freedom, and then elliptic regularity and normalized gluing yield plotwise smoothness. The theorem holds at every finite parameter regularity $C^d$ and on reduced singular parameter spaces with ambient extensions. For a Higgs deformation $\eta$, the normalized metric variation satisfies $L_hs=-\mathcal S_h(\eta)$ and $s=-G_h\mathcal S_h(\eta)$ up to an independent rank-one determinant term for $\mathrm{GL}_r$. This computes the plotwise differential and recovers the classical comparison. Locally split, constant-type polystable families admit smooth harmonic metrics. Real-analytic examples show general polystable families may have neither continuous harmonic metrics nor relative harmonic filtrations and may lie outside every $C^d$ extension-generated locus. In one example a singular harmonic reduction produces a continuous adjoint Higgs field and a flat family with semisimple slices. This defines a weak $C^0$ operator-level harmonic mediator, strictly larger than the metric-regular one, whose endpoint images after finite extension completion and stackification satisfy $\mathscr M_{\mathrm{Dol},0}^{\mathrm{wk}\mathcal H}(X)\simeq\mathscr M_{\mathrm{dR},0}^{\mathrm{wk}\mathcal H}(X)$. We characterize the extension-generated stack by relative harmonic filtrations, develop their obstruction theory, analyze the loss of extension data under heat flow, and construct the smooth Hodge $\lambda$-family on the stable locus.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

16 extracted references · 5 linked inside Pith

[1]

Azam and S

M. Azam and S. Rayan,Moduli stacks of quiver connections and non-Abelian Hodge theory, arXiv:2512.12188 (2025)

arXiv 2025
[2]

Azam and S

M. Azam and S. Rayan,A diffeological perspective on non-Abelian Hodge theory, arXiv:2606.16772 (2026)

arXiv 2026
[3]

Corlette,Flat G-bundles with canonical metrics, J

K. Corlette,Flat G-bundles with canonical metrics, J. Differential Geom.28 (1988), no. 3, 361–382

1988
[4]

S. K. Donaldson,Twisted harmonic maps and the self-duality equations, Proc. London Math. Soc. (3)55 (1987), no. 1, 127–131

1987
[5]

N. J. Hitchin,The self-duality equations on a Riemann surface, Proc. London Math. Soc. (3)55 (1987), no. 1, 59–126

1987
[6]

N. J. Hitchin,A universal Higgs bundle moduli space, arXiv:2603.21712 (2026)

arXiv 2026
[7]

T. Hu, M. Shi, R. Sun, and K. Zuo, Kodaira–Spencer map on the Hitchin–Simpson correspondence, arXiv:2511.14272 (2025)

arXiv 2025
[8]

T. Hu, R. Sun, J. Yang, and K. Zuo,Higher order isomonodromic deformation of Higgs bundles and a characterization of the non-abelian Noether–Lefschetz locus, arXiv:2606.05794 (2026)

arXiv 2026

Show all 16 references

[9]

T. Hu, R. Sun, J. Yang, and K. Zuo,Isomonodromic deformations, C∗-actions, and characterization of non-abelian Noether–Lefschetz loci on Dolbeault moduli spaces, arXiv:2606.18768 (2026)

2026 arXiv
[10]

Kim and G

S. Kim and G. Wilkin,Analytic convergence of harmonic metrics for parabolic Higgs bundles, J. Geom. Phys. 127 (2018), 55–67

2018
[11]

Kryczka, Y

J. Kryczka, Y. Tanaka, and S.-T. Yau,Lagrangian correspondences of nonabelian Hodge type and shifted twistor structures, arXiv:2606.30637 (2026)

2026 arXiv
[12]

C. T. Simpson, Constructing variations of Hodge structure using Yang–Mills theory and applications to uniformization, J. Amer. Math. Soc.1 (1988), no. 4, 867–918

1988
[13]

C. T. Simpson,Higgs bundles and local systems, Inst. Hautes Études Sci. Publ. Math.75 (1992), 5–95

1992
[14]

C. T. Simpson,Moduli of representations of the fundamental group of a smooth projective variety. II, Inst. Hautes Études Sci. Publ. Math.80 (1994), 5–79

1994
[15]

Wilkin,Morse theory for the space of Higgs bundles, Comm

G. Wilkin,Morse theory for the space of Higgs bundles, Comm. Anal. Geom.16 (2008), no. 2, 283–332

2008
[16]

Uhlenbeck and S.-T

K. Uhlenbeck and S.-T. Yau,On the existence of Hermitian–Yang–Mills connections in stable vector bundles, Comm. Pure Appl. Math.39 (1986), suppl., S257–S293. DIFFEOLOGICAL NON-ABELIAN HODGE THEORY: METRICS AND DEFORMATIONS 103 Email address: mahmud.azam@usask.ca Email address:...

1986

Pith tools

Reviewed August 1, 2026 · model on record in the stance chip above.

[1] [1]

Azam and S

M. Azam and S. Rayan,Moduli stacks of quiver connections and non-Abelian Hodge theory, arXiv:2512.12188 (2025)

arXiv 2025

[2] [2]

Azam and S

M. Azam and S. Rayan,A diffeological perspective on non-Abelian Hodge theory, arXiv:2606.16772 (2026)

arXiv 2026

[3] [3]

Corlette,Flat G-bundles with canonical metrics, J

K. Corlette,Flat G-bundles with canonical metrics, J. Differential Geom.28 (1988), no. 3, 361–382

1988

[4] [4]

S. K. Donaldson,Twisted harmonic maps and the self-duality equations, Proc. London Math. Soc. (3)55 (1987), no. 1, 127–131

1987

[5] [5]

N. J. Hitchin,The self-duality equations on a Riemann surface, Proc. London Math. Soc. (3)55 (1987), no. 1, 59–126

1987

[6] [6]

N. J. Hitchin,A universal Higgs bundle moduli space, arXiv:2603.21712 (2026)

arXiv 2026

[7] [7]

T. Hu, M. Shi, R. Sun, and K. Zuo, Kodaira–Spencer map on the Hitchin–Simpson correspondence, arXiv:2511.14272 (2025)

arXiv 2025

[8] [8]

T. Hu, R. Sun, J. Yang, and K. Zuo,Higher order isomonodromic deformation of Higgs bundles and a characterization of the non-abelian Noether–Lefschetz locus, arXiv:2606.05794 (2026)

arXiv 2026