Connections, metrics and Higgs fields on complex fiber bundles

Mao Sheng; Nianzi Li

REVIEW 3 major objections 5 minor 1 cited by

Connections, metrics and Higgs fields on complex fiber bundles

T0 review · 3 major / 5 minor · reviewed 2026-08-02 · deepseek-v4-flash

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

Pith's one-line read The extension class of any holomorphic fibration is represented by a canonical curvature tensor, and this yields a faithful correspondence between nonlinear flat bundles of Kähler type and nonlinear Higgs bundles.

desk verdict Solid core, soft edges: clean Atiyah-class curvature and nonlinear Riemann-Hilbert theorems, but the flagship faithful functor is proved only on a restricted morphism category and the VHS section is a proof sketch; deserves serious refereeing. read the letter →

arxiv 2602.13838 v2 pith:Q3OMLBU7 submitted 2026-02-14 math.DG math.AG

Nianzi Li , Mao Sheng This is my paper · ORCID

classification math.DGmath.AG MSC 14D0732Q1553C0714D21

keywords holomorphicfibrationAtiyahclassnonlinearflatbundlesHiggsharmonicmetricsKählertypenonabelianHodgecorrespondencetwistedSimpsonmechanism

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

This paper establishes that the cohomology class obstructing holomorphic connections on an arbitrary holomorphic fibration—not just principal bundles—equals the class of a canonically constructed ∂̄-closed tensor made from any pure complex connection. From this it derives a nonlinear version of the classical theorem characterizing when a holomorphic fibre bundle over a compact Riemann surface admits a flat connection, in terms of degrees of the adjoint bundle and torsion of a characteristic class. It then builds a faithful functor from reductive flat bundles of Kähler type over a compact Kähler base to nonlinear Higgs bundles, using harmonic metrics obtained from the classical harmonic-map existence theorem. In the last part it introduces the twisted Simpson mechanism and shows that the variation of nonabelian Hodge structure is a nonlinear harmonic bundle in the rank-one and semisimple cases.

What carries the argument

The central object is the extension class A(X)∈H^1(X,f^*Ω_S⊗T_{X/S}) of the relative tangent sequence, together with the tensor R associated to a pure complex connection: locally R is (∂̄_sΓ + ∂̄_zΓ) ds⊗∂_z and is proven ∂̄_X-closed with class A(X). This replaces a sheaf-cohomology obstruction with a differential-geometric curvature for all fibrations. The second load-bearing mechanism is the (twisted) harmonic-metric mechanism: a fibrewise Kähler metric whose stabilizer K is a compact real form yields a Chern connection, and the harmonic-map existence theorem produces a canonical metric; from the flat connection and this metric one forms θ=(∂−∂^Ch)/2 and a deformed ∂̄-operator, whose vanish

What would settle it

Compute, for a concrete holomorphic fibration with nonzero extension class (e.g., a non-isotrivial elliptic fibration over a curve), the tensor R from a pure complex connection and test whether its Dolbeault class equals the Čech class A(X); if the two classes differ, Theorem 1.3 fails. Alternatively, find two G-harmonic fibrewise Kähler metrics on the same reductive flat bundle of Kähler type that yield non-isomorphic nonlinear Higgs bundles, which would falsify Proposition 5.18's independence claim.

Watch

Extended reading notes

Core claim

For a holomorphic fibration f:X→S, the short exact sequence of tangent bundles defines an extension class A(X) in H^1(X,f^*Ω_S⊗T_{X/S}); it vanishes exactly when f admits a holomorphic connection. The paper proves that for any pure complex connection the local curvature expression assembles into a global ∂̄-closed tensor R whose Dolbeault class equals A(X), recovering the classical statement that the Atiyah class is the curvature class of a connection. When the connection is relatively holomorphic, R is simply the (1,1)-curvature of the connection, and the Kodaira-Spencer map vanishes. Using this, a holomorphic fibre bundle over a compact Riemann surface with reductive structure group admits

Load-bearing premise

The load-bearing assumption is that the fibre Y carries a Kähler metric ω_Y whose stabilizer K=Stab_G(ω_Y) is a compact real form of the automorphism group G; the existence of such a metric is what supplies the harmonic map and the Chern connection, and the paper provides examples but no general criterion for when one exists.

Editorial extensions

If this is right

The vanishing of A(X) — equivalently existence of a holomorphic connection on the fibration — is now detected by the cohomology class of a curvature tensor computed from any pure complex connection, making it checkable by differential-geometric data.
For holomorphic fibre bundles over compact Riemann surfaces with reductive structure group, existence of a flat (hence holomorphic) connection is equivalent to the degree-zero condition on the Remak summands of the adjoint bundle together with torsion of the characteristic class c(P).
Complete nonlinear flat bundles over a connected complex manifold are equivalent to representations of the fundamental group of the base into the automorphism group of the fibre.
Every reductive nonlinear flat bundle of Kähler type over a compact Kähler base carries a harmonic fibrewise Kähler metric, independent of choices, producing a well-defined nonlinear Higgs bundle and a faithful functor between the two categories.
In the rank-one and semisimple cases, the flat bundle underlying a variation of nonabelian Hodge structure and its associated graded Higgs bundle are related by the twisted Simpson mechanism, making the variation a nonlinear harmonic bundle.

Reading between the lines

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

Inference: if the curvature-class representation is natural under pullbacks, the whole construction should descend to moduli stacks of fibrations; a direct check of base-change compatibility would sharpen the paper's claims.
Inference: the faithful functor is not shown to be full; proving fullness would require a nonlinear analogue of the Hermitian–Yang–Mills equation, a natural next step the paper points to but does not resolve.
Inference: the dependence of Theorem 6.29 on a smooth isomorphism between de Rham and Dolbeault moduli spaces suggests the result will extend to arbitrary reductive G exactly when such an isomorphism exists on a suitable Zariski-dense smooth locus; testing on non-semisimple groups would delimit the mechanism.
Inference: the completeness condition in the nonlinear Riemann-Hilbert correspondence may be unnecessary for compact fibres but is essential for noncompact ones, as the blow-up example shows; extending the equivalence to incomplete connections with extra regularity would be a testable refinement.

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

Solid core, soft edges. The curvature representative of the Atiyah class for arbitrary holomorphic fibrations (Thm 1.3) is a real result: the proof via a Cech-Dolbeault double complex is self-contained and correct, and it genuinely extends Atiyah's principal-bundle picture. The nonlinear Riemann-Hilbert equivalence (Thm 3.18/1.8) is also clean and complete as stated, with the completeness hypothesis doing exactly the work needed. The harmonic-metric construction in Section 5 is credible: it imports Donaldson-Corlette in the right way and produces a nonlinear Higgs bundle from the flat bundle. I would cite the curvature theorem and the Riemann-Hilbert correspondence without hesitation.

Where I part company with the introduction is Theorem 1.11. As stated, it promises a faithful functor on the category of reductive Kahler-type flat bundles with all morphisms of complete flat bundles. Proposition 5.19 proves faithfulness only for morphisms induced by alpha-equivariant maps phi with alpha(K1) subset K2. That is a real restriction: a general morphism preserving horizontal distributions need not be of that form, and Remark 5.20 concedes the full-category version is expected rather than proved. The stress-test note is right on this point. This is not a spoiled theorem — the restricted statement is proved — but the abstract and introduction overstate what the proof delivers.

The last section is noticeably weaker. Theorem 6.29 is stated for semisimple or C* and says the flat and Higgs bundles are related by the twisted Simpson mechanism. The proof is a sketch: it relies on Lemma 6.28's smooth isomorphism from an imported reference, on unproved pullback compatibility, and on a uniqueness claim for a local holomorphic map to Teichmuller space. I would not call this a demonstrated error, but it is not a proof at the same standard as Sections 2-5.

Net: the paper deserves a serious referee. It is long, specialized, and the introduction needs an honest repair — either prove the full morphism statement or state the restricted theorem accurately. I would send it out, and ask the referee to focus on the gap between Prop 5.19 and Thm 1.11, and on making the VHS section honest about its dependencies. Someone working on Atiyah classes or nonlinear Hodge theory will get real value from the core sections; I would bring the first half to a reading group. I would not yet rely on the full Theorem 1.11 or Theorem 6.29 as stated.

Referee Report

3 major / 5 minor

Summary. The paper develops nonlinear analogues, for arbitrary holomorphic fibrations f:X→S, of Atiyah's curvature interpretation of the extension class, of Weil's flat-bundle criterion over Riemann surfaces, and of parts of the nonabelian Hodge correspondence. Theorem 1.3 canonically associates to any pure complex connection a ∂̄_X-closed tensor R with class A(X)∈H^1(X,f^*Ω_S⊗T_{X/S}); Theorem 1.8 establishes an equivalence between complete nonlinear flat bundles and representations of π_1(S) into Aut(Y); Corollary 1.5 gives a nonlinear Weil-type criterion for associated bundles over a compact Riemann surface. The paper then introduces harmonic fiberwise Kähler metrics and, via the Donaldson–Corlette theorem, constructs a faithful functor from reductive Kähler-type nonlinear flat bundles to nonlinear Higgs bundles (Theorem 1.11/Proposition 5.19). Finally, a twisted Simpson mechanism is introduced and applied to variations of nonabelian Hodge structure, with Theorem 6.29 claiming that, for G=C^* or semisimple G, the relative de Rham and Dolbeault moduli spaces are related by this mechanism.

Significance. The Atiyah-class half of the paper is a genuine and useful generalization of Atiyah's classical result: Theorem 1.3 is proved in a self-contained Čech–Dolbeault manner and gives a concrete curvature representative of the extension class for arbitrary fibrations. The nonlinear Riemann–Hilbert correspondence of Theorem 1.8 is also clean and, together with Proposition 2.23, yields a plausible nonlinear Weil criterion. If the categorical and moduli-theoretic claims were fully established, the paper would open a substantial new direction in nonabelian Hodge theory. However, the headline faithful-functor statement is stronger than the proposition actually proved, and the final section depends on unproved compatibility statements and on substantial imported results. The significance is therefore conditional on repairing the overclaims.

major comments (3)

[§1, Theorem 1.11; §5.3, Proposition 5.19] The abstract and Theorem 1.11 state that there is a faithful functor from the category of nonlinear flat bundles reductive of Kähler type over S, as defined by Definition 1.6, to nonlinear Higgs bundles. Proposition 5.19, which is cited as the proof, only constructs the functor on the restricted category RFB(S) whose morphisms are maps of the form [\tilde{s},y]_{ρ1} ↦ [\tilde{s},φ(y)]_{ρ2}, where φ is α-equivariant for a Lie homomorphism α:G1→G2 satisfying α(K1)⊂K2. The faithfulness proof uses this extra structure essentially: it pushes the harmonic map h1 forward by the totally geodesic map \tilde{α}, then compares with the chosen harmonic map using an element of the centralizer. A general morphism in Definition 1.6 need not be induced by such an α-equivariant φ, need not intertwine the harmonic-metric conjugates, and need not be holomorphic for the new complex structures ∂̄_{ω_i}; henc
[§6, Theorem 6.29; §6.3; Lemma 6.28] The proof of Theorem 6.29 for semisimple G says: 'The result follows from the universal case as carried out in §6.3' because 'the (twisted) Simpson mechanism is compatible with pullback.' This pullback compatibility is asserted but never stated or proved. The universal case in §6.3 also relies on Lemma 6.28, whose proof uses that the map F is a real-analytic diffeomorphism by [CTW25, Th. 4.23] and the real-analytic inverse function theorem; but no argument is given that the resulting smooth isomorphism between M_dR and M_Dol can be chosen to interact correctly with the twisting map β and the fiberwise metric. Consequently, the reconstruction claim of Theorem 6.29 is not fully justified. The authors should either prove the needed compatibility or make the dependence on it an explicit additional hypothesis.
[§5.1–5.3, Assumption 1.9/5.13] Assumption 1.9 is load-bearing for the whole harmonic-metric construction: the paper requires a Kähler metric ω_Y on the fiber whose stabilizer K=Stab_G(ω_Y) is a compact real form of G. The paper gives the cscK example (Example 5.25) but no general criterion for the existence of such an ω_Y. Since this condition is part of the definition of the category in Theorem 1.11, the theorem is conditional on a possibly very restrictive geometric condition. The introduction should state this more prominently, and ideally the paper should discuss what is known or conjectured about existence of such ω_Y beyond the cscK case. This does not invalidate the conditional statements, but it affects the claimed scope of the 'nonlinear Hodge correspondence.'

minor comments (5)

[§6.1, Equation (6.1)] The definition of ¯θ_J in (6.1) would benefit from an explicit statement of the domains and codomains of J^A_{X/S}, J^B_{X/S}, and of the projection pr_{T^B_{X/S}}. As written, the composition is only readable after inferring the intended fiberwise identifications.
[§3.15, Lemma 3.15] The argument that β(u) lies in L_{x0}∩X_{s1}, which is discrete, is correct but terse; adding one sentence explaining that the transverse foliation intersects fibers discretely would improve readability.
[§2.1, Definition 2.3] The notation f^*TX/S is used both for a sheaf of smooth fiberwise holomorphic sections and, later, as a vector bundle over X. This dual use is potentially confusing; a remark distinguishing the sheaf and bundle interpretations would help.
[Corollary 3.21] The phrase 'Remak decomposition' should be checked; standard spelling is 'Remak' or 'Remak decomposition' depending on convention. More importantly, the converse part of the statement depends on finite-dimensionality of H^0(Y,TY), which is stated in the hypothesis, so the statement is internally fine but should perhaps be highlighted as the key restriction.
[§6.4, Definition 6.30] The sentence 'A choice of a twisting map and a (β-twisted) harmonic metric is not part of the data' conflicts slightly with the preceding theorem, which produces a suitable twisting map and metric. Consider clarifying that the object is the pair (flat bundle, Higgs bundle) modulo the existence of some such choice, rather than including the choice in the data.

Circularity Check

0 steps flagged · score 2.0 of 10

No significant circularity: Theorems 1.3, 1.8, and 5.18/5.19 are derived from self-contained cohomology computations, explicit parallel-transport constructions, and classical harmonic-map input. The main flagged issue is an overstatement of the morphism category in Theorem 1.11 relative to Proposition 5.19 (conceded in Remark 5.20); this is a correctness gap rather than a circular reduction.

full rationale

I walked the derivation chain and found no equation that reduces to its own output by construction, and no fitted quantity renamed as a prediction. Theorem 1.3/3.3 is proved by writing the Atiyah class as a Čech cocycle c_ab = σ_a − σ_b and the curvature tensor R locally as ¯∂_X(Γ ds ⊗ ∂), then exhibiting γ = ∇^{1,0} − σ_a with d_tot γ = c − R; this is a direct double-complex cohomology argument, not a circular one. Theorem 1.8/3.18 constructs the Riemann–Hilbert equivalence explicitly from the quotient ( ̃S × Y)/π_1 and from parallel transport of complete flat connections; the proof does not import the conclusion. Proposition 5.18/5.19 uses the Donaldson–Corlette harmonic map h: ̃S → G/K, the induced K-reduction P_K, and the standard decomposition A = A_K + ψ to obtain θ = ψ^{1,0} and ¯∂ = A_K^{0,1}; the faithfulness proof is an injectivity argument on morphisms and does not reduce a prediction to a fit. The self-citations [She25a, She25b, FS25] overlap with the authors, but they are used as motivation or as external ingredients; FS25 supplies the nonabelian Kodaira–Spencer map θ_KS in Section 6, which is a dependency, not a reconstruction of this paper's claims from themselves. The real limitation is categorical: Proposition 5.19 constructs the faithful functor only on the subcategory RFB(S) whose morphisms are induced by α-equivariant maps φ with α(K_1) ⊂ K_2, whereas Theorem 1.11 and the abstract claim the functor for the full category of nonlinear flat bundles reductive of Kähler type under Definition 1.6. Remark 5.20 explicitly concedes that faithfulness for the full subcategory of CFB(S) is only expected. That is an overstatement/missing proof, not circularity. Accordingly, the circularity score is low: the central derivations are self-contained, and no step is forced by definition or by a self-citation chain.

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

No numeric free parameters are fitted; the paper's choices are geometric structures such as Kähler metrics and twisting maps. The central claims rest on the compact-real-form assumption, imported harmonic-map theorems, and recent moduli-space results, several of which involve the same author circle. The new objects introduced, such as nonlinear harmonic bundles and twisting maps, are definitions with proof obligations rather than unsupported entities.

assumptions (5)

domain assumption Donaldson-Corlette existence of equivariant harmonic maps to G/K for reductive representations
Imported in §5.3 (Theorem 5.15) and used to build G-harmonic metrics; without it the faithful functor of Propositions 5.18-5.19 does not exist.
ad hoc to paper Assumption 1.9: there exists a Kähler metric ω_Y on Y with K=Stab_G(ω_Y) a compact real form of G
Defines the class of Kähler-type nonlinear flat bundles; not proven for general Y, only exemplified by cscK manifolds (Example 5.25).
domain assumption Existence and properties of the joint moduli space of G-Higgs bundles from [CTW25]
Used in §6.3 to identify fibers of M(G), the relative Kähler form ω0, and the nonabelian Kodaira-Spencer map; this is heavy external machinery.
domain assumption FS25 Theorem 1.2: θ_KS coincides with the residual action of the nonabelian Gauss-Manin connection
Used in Theorem 6.29; the paper depends on this overlapping-author result without independent verification in this preprint.
domain assumption Lemma 6.28: M_dR and M_Dol are isomorphic as smooth fiber bundles via the real-analytic nonabelian Hodge map
Necessary for the twisted Simpson mechanism to compare the two moduli spaces over the same smooth fiber bundle; relies on [CTW25, Thm 4.23].

how reviews work

0 comments

Cite this review

Pith. "Pith review of Connections, metrics and Higgs fields on complex fiber bundles." pith.science (2026). https://pith.science/paper/Q3OMLBU7

@misc{pith2026260213838,
  author       = {Pith},
  title        = {Pith review of: Connections, metrics and Higgs fields on complex fiber bundles},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/Q3OMLBU7}},
  note         = {Machine review of arXiv:2602.13838}
}

read the original abstract

We give a representation of the extension class associated to a holomorphic fibration by curvature, generalizing the work of Atiyah on holomorphic principal bundles in a natural way. As an application, we obtain a nonlinear analogue of the classical result of Weil on characterizing the existence of flat connections on holomorphic vector bundles over compact Riemann surfaces. We further establish a faithful functor from the category of nonlinear flat bundles reductive of K\"ahler type to the category of nonlinear Higgs bundles over the same base, which is assumed to be a compact complex manifold of K\"ahler type. Finally, we establish a notion of nonlinear harmonic bundle and prove that the variation of nonabelian Hodge structure is a nonlinear harmonic bundle in the rank one case and in the semisimple case.

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

Nonlinear Hodge correspondence for morphisms
math.DG 2026-07 accept novelty 8.0 of 10

In nonlinear harmonic bundles, a section is flat if and only if it is a Higgs section with vanishing degree; the same equivalence extends to sub-fibrations and morphisms.

Reference graph

Works this paper leans on

6 extracted references · 1 linked inside Pith · cited by 1 Pith paper

[1]

II, Bull

[AB03] Hassan Azad and Indranil Biswas,On holomorphic principal bundles over a compact Riemann surface admitting a flat connection. II, Bull. London Math. Soc.35(2003), no. 4, 440–444. MR 1978996 [Ati57] M. F. Atiyah,Complex analytic connections in fibre bundles, Trans. Amer. Math. Soc.85(1957), 181–207. MR 86359 [BB17] Robert J. Berman and Bo Berndtsson,...

2003
[590]

MR 3309256 [UY86] Karen Uhlenbeck and Shing-Tung Yau,On the existence of Hermitian-Yang-Mills connections in stable vector bundles, Comm

MR 111062 [Son15] Stephen Bruce Sontz,Principal bundles, Universitext, Springer, Cham, 2015, The classical case. MR 3309256 [UY86] Karen Uhlenbeck and Shing-Tung Yau,On the existence of Hermitian-Yang-Mills connections in stable vector bundles, Comm. Pure Appl. Math.39(1986), S257–S293. MR 861491 [Wei38] Andr´ e Weil,G´ en´ eralisation des fonctions ab´ e...

2015
[1972]

Korevaar and Richard M

MR 355886 [KS93] Nicholas J. Korevaar and Richard M. Schoen,Sobolev spaces and harmonic maps for metric space targets, Comm. Anal. Geom.1(1993), no. 3-4, 561–659. MR 1266480 [KS97] ,Global existence theorems for harmonic maps to non-locally compact spaces, Comm. Anal. Geom.5(1997), no. 2, 333–387. MR 1483983 [LN02] Alcides Lins Neto,Some examples for the ...

arXiv 1993
[1985]

Differential Geom.56(2000), no

MR 781344 [Che00] Xiuxiong Chen,The space of K¨ ahler metrics, J. Differential Geom.56(2000), no. 2, 189–234. MR 1863016 [Che12] Ting Chen,The associated map of the nonabelian Gauss-Manin connection, Cent. Eur. J. Math. 10(2012), no. 4, 1407–1421. MR 2925612 [CK21] Diego Corro and Jan-Bernhard Kordaß,Short survey on the existence of slices for the space o...

arXiv 2000
[2013]

Math.190 (2012), no

[Sch12] Georg Schumacher,Positivity of relative canonical bundles and applications, Invent. Math.190 (2012), no. 1, 1–56. MR 2969273 [She25a] Mao Sheng,Nonlinear harmonic bundles, arXiv e-prints (2025), arXiv:2512.04809. [She25b] Mao Sheng,Nonlinear Hodge correspondence in positive characteristic, arXiv e-prints (2025), arXiv:2510.05578. [Sim88] Carlos T....

arXiv 2012
[2017]

MR 3674984 [NS65] M. S. Narasimhan and C. S. Seshadri,Stable and unitary vector bundles on a compact Riemann surface, Ann. of Math. (2)82(1965), 540–567. MR 184252 [OV07] A. Ogus and V. Vologodsky,Nonabelian Hodge theory in characteristicp, Publ. Math. Inst. Hautes ´Etudes Sci. (2007), no. 106, 1–138. MR 2373230 [Ram75] A. Ramanathan,Stable principal bund...

1965

Pith tools

Reviewed August 2, 2026 · model on record in the stance chip above.

[1] [1]

II, Bull

[AB03] Hassan Azad and Indranil Biswas,On holomorphic principal bundles over a compact Riemann surface admitting a flat connection. II, Bull. London Math. Soc.35(2003), no. 4, 440–444. MR 1978996 [Ati57] M. F. Atiyah,Complex analytic connections in fibre bundles, Trans. Amer. Math. Soc.85(1957), 181–207. MR 86359 [BB17] Robert J. Berman and Bo Berndtsson,...

2003

[2] [590]

MR 3309256 [UY86] Karen Uhlenbeck and Shing-Tung Yau,On the existence of Hermitian-Yang-Mills connections in stable vector bundles, Comm

MR 111062 [Son15] Stephen Bruce Sontz,Principal bundles, Universitext, Springer, Cham, 2015, The classical case. MR 3309256 [UY86] Karen Uhlenbeck and Shing-Tung Yau,On the existence of Hermitian-Yang-Mills connections in stable vector bundles, Comm. Pure Appl. Math.39(1986), S257–S293. MR 861491 [Wei38] Andr´ e Weil,G´ en´ eralisation des fonctions ab´ e...

2015

[3] [1972]

Korevaar and Richard M

MR 355886 [KS93] Nicholas J. Korevaar and Richard M. Schoen,Sobolev spaces and harmonic maps for metric space targets, Comm. Anal. Geom.1(1993), no. 3-4, 561–659. MR 1266480 [KS97] ,Global existence theorems for harmonic maps to non-locally compact spaces, Comm. Anal. Geom.5(1997), no. 2, 333–387. MR 1483983 [LN02] Alcides Lins Neto,Some examples for the ...

arXiv 1993

[4] [1985]

Differential Geom.56(2000), no

MR 781344 [Che00] Xiuxiong Chen,The space of K¨ ahler metrics, J. Differential Geom.56(2000), no. 2, 189–234. MR 1863016 [Che12] Ting Chen,The associated map of the nonabelian Gauss-Manin connection, Cent. Eur. J. Math. 10(2012), no. 4, 1407–1421. MR 2925612 [CK21] Diego Corro and Jan-Bernhard Kordaß,Short survey on the existence of slices for the space o...

arXiv 2000

[5] [2013]

Math.190 (2012), no

[Sch12] Georg Schumacher,Positivity of relative canonical bundles and applications, Invent. Math.190 (2012), no. 1, 1–56. MR 2969273 [She25a] Mao Sheng,Nonlinear harmonic bundles, arXiv e-prints (2025), arXiv:2512.04809. [She25b] Mao Sheng,Nonlinear Hodge correspondence in positive characteristic, arXiv e-prints (2025), arXiv:2510.05578. [Sim88] Carlos T....

arXiv 2012

[6] [2017]

MR 3674984 [NS65] M. S. Narasimhan and C. S. Seshadri,Stable and unitary vector bundles on a compact Riemann surface, Ann. of Math. (2)82(1965), 540–567. MR 184252 [OV07] A. Ogus and V. Vologodsky,Nonabelian Hodge theory in characteristicp, Publ. Math. Inst. Hautes ´Etudes Sci. (2007), no. 106, 1–138. MR 2373230 [Ram75] A. Ramanathan,Stable principal bund...

1965