Alphabeta Math
LemmaStatement: Literature-sourcedProof: AI-adaptedPipeline-generatedprecheck passverified 2026-11-05 (gpt-6.1-sol)
How statement and proof provenance work

The first chip identifies the source of the statement or construction; the second identifies the source of its local proof or verification.

  • Literature-sourced: the exact statement appears in a cited source; only wording and notation differ.
  • AI-adapted: a semantically identical restatement of literature-sourced material, modulo indexing, notation, and boundary cases adopted by the library.
  • AI-generated: a genuinely novel statement formulated by AI, with no source for the claim itself.

These labels describe origin, not correctness: citations and verification chips remain separate evidence.

Sharp monodromy-weight growth of the Hodge metric

Statement

Let a polarized real variation of pure Hodge structure on the punctured disk have unipotent marking action T=exp⁡N, and write t=exp⁡(2πiz), z=x+iy. A real variation here means a flat real vector bundle with holomorphic Hodge filtration, real bilinear polarization and Griffiths transversality, without requiring an integral lattice. Let W(N) be the zero-centred monodromy weight filtration. A grading H of W(N) is a real semisimple endomorphism of the flat real vector space with integral eigenvalues and decomposition VR=⨁lEl(H) satisfying Wj(N)=⨁l≤jEl(H). Assume [H,N]=−2N, and choose a fixed positive metric making its complexified eigenspaces orthogonal. On a bounded interval of x and for y sufficiently large, there exist c,C>0 such that, for all flat vectors v,

c∑lyl∣∣vl∣∣2≤hx+iy(v,v)≤C∑lyl∣∣vl∣∣2,

where v=∑lvl is the H eigendecomposition. For every nonzero flat vector v,v∈Wl(N)∖Wl−1(N) if and only if its squared Hodge norm is bounded above and below by positive multiples of yl. No integrality or rationality of N is needed.

Facts & Assumptions

Given: The analytic polarized real variation, N,H and fixed positive metric in the Statement.

[F1]

The horizontal curvature argument gives ||A||≤C/y for θ=A dz (Horizontal period maps have bounded metric differential). Its finite connection/trace/max-principle proof requires no lattice, so applies to the real data specified here.

[F2]

The canonical nilpotent weight filtration has a finite primitive string decomposition (The nilpotent Lefschetz filtration is canonical and self-dual). A homogeneous grading H with N-degree −2 makes that decomposition H-homogeneous.

Proof

technique · direct
1.1F1F2givenalgebra

On the universal cover write the Hodge splitting as Ep=Fp∩F‾k−p and the flat connection as ∇=D+A dz+A†dzˉ, where D preserves the positive Hodge metric h and A lowers Hodge degree by one. These are the actual-variation identities proved in the horizontal-period-map supplier. If v is flat, metric compatibility and ∇v=0 give ∂zh(v,v)=−2h(Av,v) and ∂z∂zˉh(v,v)=2(∥Av∥2+∥A†v∥2).

2.1F1F2step 1.1givenalgebra

For v≠0, set u=log⁡h(v,v). The two derivative identities and Cauchy–Schwarz show ∂z∂zˉu≥0 and ∣du∣≤C∥A∥. If v is monodromy invariant, u is 1-periodic in x. The horizontal bound ∥A∥≤C/y therefore gives ∣du∣≤C′/y. The period average u0(y)=∫01u(x,y) dx satisfies u0′′≥0 and ∣u0′∣≤C′/y; hence u0′ increases to 0, is nonpositive, and u0 is bounded above. The x-derivative bound then gives ∣u−u0∣≤C′/y on a period. Thus every invariant flat vector has uniformly bounded Hodge norm near the puncture.

3.1step 2.1givenalgebra

If two polarized complex variations have isomorphic flat bundles, their Hodge metrics are uniformly mutually bounded near the puncture: apply step 2.1 to the invariant flat isomorphism in the Hom variation and to its inverse. The Hom polarization gives the Hilbert–Schmidt metric, which dominates the operator norm, so these two bounds compare the metrics on every vector.

4.1F1step 1.1step 2.1step 3.1algebra

We construct the auxiliary metric used in step 3.1 explicitly. A complex polarized variation here means a flat complex bundle with a smooth grading Ep, a flat nondegenerate Hermitian form B, holomorphic filtration Fp=⨁a≥pEa, transverse connection, and positive metric h∣Ep=(−1)k−pB∣Ep for a fixed integer k. This auxiliary category need not have a real structure or a lattice. The calculation in [F1] applies to it: differentiating B(Ep,Ep−1)=0 gives the same adjacent-adjoint identity, and flatness gives the same DzˉA=0 and FD=−[A,A†]. Thus the derivative identities in step 1.1 and the invariant-vector argument in step 2.1 hold in this category. The complexification of the original real variation is in it with B(u,v)=ikQ(u,vˉ). Hom gradings are by degree difference; their flat trace Hermitian form has the corresponding alternating signs and positive Hilbert–Schmidt metric. Their Higgs operator acts by f↦A2f−fA1, of norm at most ∥A1∥+∥A2∥. Consequently step 3.1 applies whenever both original and auxiliary Higgs norms are O(1/y), without requiring their flat identification to preserve either filtration or polarization.

5.1step 4.1constructalgebra

On a complex two-dimensional space with basis a,b, let Ya=b, Yb=0, Ha=a, Hb=−b, and let B have matrix (0110). Put F1=Ca, F0=C2 and F(z)=e−izYF. At z=iy its two B-orthogonal lines are C(a+yb) and C(a−yb), with B respectively positive and negative; this gives a polarized complex variation of weight one. Its positive metric in the basis a,b is diag⁡(y,y−1). Real translation acts by e−ixY, a B-isometry, since Y is B-self-adjoint. Holomorphicity and transversality follow by differentiating the displayed line, and its logarithmic monodromy is −iY. Differentiating the two normalized orthogonal lines gives Higgs norm a fixed multiple of 1/y.

6.1F2step 5.1constructalgebra

On Sm=Sym⁡mC2, the monomial basis am−jbj is orthogonal at iy and has squared norms cmjym−2j, with fixed cmj>0 from the symmetrized tensor metric. Its Higgs operator is the sum of the m tensor Higgs operators, so is O(1/y). Its Hodge indices may be shifted by tensoring a constant one-dimensional complex Hodge structure; choosing that line's Hermitian sign gives any common total weight k without changing the positive metric. This changes neither the connection nor its Higgs bound. A homogeneous primitive-string decomposition for the supplied N,H identifies each string with Sm: scale the monomial vectors successively so that −iY becomes exactly N. Therefore their finite direct sum is a complex polarized auxiliary variation on the same flat bundle, with metric ∑lyl∥vl∥∗2 at iy, for a fixed positive H-orthogonal norm ∥⋅∥∗.

7.1F1F2step 3.1step 4.1step 6.1algebra

Apply steps 2.1–4.1 to the invariant flat identification of this auxiliary bundle with the original bundle and to its inverse. Both Higgs bounds have been verified, so their two Hodge metrics are uniformly equivalent. On a bounded interval of x, the auxiliary metric is obtained by the flat factor e−xN; conjugation by yH/2 gives yH/2e−xNy−H/2=e−xN/y, uniformly bounded with bounded inverse for y≥1. All fixed positive H-orthogonal norms are equivalent with constants independent of y. This proves the two-sided estimate uniformly in the stated strip, for all vectors simultaneously.

8.1F2step 7.1algebra∎

In the positive sum ∑lyl∥vl∥2 for v≠0, its largest nonzero index l determines its order above and below. The grading identity Wl=⨁a≤lEa(H) identifies this index with v∈Wl∖Wl−1. Thus the asserted equivalence follows. The construction used finite strings, real polarized data and an auxiliary complex Hermitian variation, with no integrality assumption on N.

Source-reading limits

Sabbah–Schnell §§132–142 (PDF pp. 60–66) were read, including the proofs of the invariant-vector and Hom comparison statements and the model argument. The model, comparison and full-space norm estimates used here are reconstructed in steps 1.1–8.1; the cited theorem is not an external proof premise. No reading of the whole preprint is claimed.

Depends on

Dependency tree · 2 levels

Sources