Alphabeta Math
LemmaStatement: AI-adaptedProof: AI-generatedPipeline-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.

A formal chiral operation glues over a pointed curve

Statement

Let V be a conformal vertex algebra of CFT type and X a smooth projective complex curve. For each N≥0, put V≤N:=⨁0≤d≤NVd for the conformal grading and use a holomorphic coordinate chart U with coordinate t to trivialize a bundle by U×V≤N. On an overlap with coordinate s=ρ(t), put ρy(u):=ρ(y+u)−ρ(y), where y=t(p) at the base point p; glue the finite-stage bundles by the holomorphic transition RV(ρy)∣V≤N from the preceding formal-coordinate covariance lemma. These transitions preserve V≤N and satisfy the cocycle by the right-action law. Let VX be the filtered union of the resulting bundles. Let ωX:=ΩX,hol1 be the holomorphic cotangent line bundle, whose local sections in coordinate t are f(t)dt, and put VXr:=ωX⊗OXVX. Give VX the left connection ∇∂t=∂t+T; on VXr the corresponding right-DX action is

(f(t)dt⊗v(t))⋅∂t=−f′(t)dt⊗v(t)−f(t)dt⊗(∂t+T)v(t).

Here T=L(−1). In coordinate t, define the regular-input operation on states by the finite delta-derivative sum χt(a,b):=∑j≥0(a(j)b)∂t2jδ(t1,t2)j!. Equivalently, its singular-OPE shorthand is ∑j≥0(a(j)b)(t1−t2)−j−1, the difference of the two formal binomial expansions. For f∈R2:=C⟦y,u⟧[u−1], define PP⁡u(∑ici(y)ui):=∑i<0ci(y)ui and use the full localized local map f(a dt1⊠b dt2)↦PP⁡u(f(y+u,y)Y(a,u)b), where a,b∈V, u=t1−t2, and y=t2. Its target is locally Q2dt2, where Q2:={∑j=0Jcj(y)u−j−1:J<∞, cj(y)∈⋃N≥0V≤N⟦y⟧}; the principal part cj(y)u−j−1dt2 represents cj(y)∂t2jδ(t1,t2)/j!. Its source consists of local sections with finite pole order along the diagonal. The full local operation obeys the three-input chiral Jacobi identity proved in the preceding right-D local-Jacobi supplier, with the Beilinson–Drinfeld convention of Chiral Algebras, Chapter 3, §3.3.3. Its coordinate-covariant local maps glue to a right DX-module operation j∗j∗(VXr⊠VXr)⟶Δ!VXr, where j:X2∖Δ↪X2, Δ:X↪X2, and Δ! is locally the finite-sum module of delta derivatives supported on the diagonal. In this chiral-algebra notation, j∗j∗E means the meromorphic localization E(∗Δ):=lim→⁡k≥0E(kΔ) for E=VXr⊠VXr, with the union over finite state stages understood; E(kΔ) allows a pole of order at most k along the diagonal. It does not mean the ordinary analytic sheaf direct image of all holomorphic sections on the complement. Locally a section is a finite sum of state pairs times holomorphic functions with a finite diagonal pole, whose formal expansion has the stated R2 form. The target coefficients are holomorphic germs along the diagonal, with the displayed Q2 as their formal expansion model. This same localization convention is used when the resulting operation is called the coordinate-independent chiral algebra. The global operation satisfies chiral Jacobi, skew-symmetry, and the vacuum identity. The local Jacobi deduction is supplied by the separate right-D local-Jacobi lemma; the present proof establishes its global gluing.

Facts & Assumptions

Given: The formal-coordinate torsor, the Aut O covariance identity, the finite OPE expansion, and the actual full-field local extension constructed in [F7]. The local Jacobi premise in the Statement is discharged by that supplier.

[F1]

Vertex algebra in the state–field convention: for every a,b∈V, a(j)b=0 for all sufficiently large j; the vacuum and creation axioms hold; and [T,Y(a,z)]=∂zY(a,z).

[F2]

Conformal vertex algebra of CFT type: L(−1)=T and V is graded in nonnegative integer weights.

[F3]

Formal distributions and fields on a vector space fixes the delta and binomial conventions. Locality determines the formal OPE coefficients: the commutator of local fields A(z),B(w) has a finite delta-derivative expansion with operator-valued coefficients Cj(w)=Res⁡z(z−w)j[A(z),B(w)]. These coefficients are fields, not states. The state-valued principal parts used here are supplied directly by [F7], rather than identified with that operator-valued commutator.

[F4]

Virasoro modes give an Aut O action with field covariance: a coordinate change conjugates the module field by RM(ρ) and sends its state and coordinate to RV(ρz)v and ρ(z), respectively.

[F5]

Smooth projective complex curve with formal-coordinate bundle supplies the finite-jet bundle charts and their Taylor-substitution cocycle. Formal-coordinate change group Aut O fixes the right-composition series product, inverse-postcomposition action on coordinate ring maps, and its free transitivity. In particular, if s=ρ(t), the centered coordinate maps satisfy τs,p=ϕρy−1τt,p; the transition from t-components to s-components is RV(ρy).

[F6]

Formal residue is invariant under coordinate change: the residue of a formal one-form is invariant under every formal-coordinate substitution.

[F7]

Full local chiral Jacobi from vertex Jacobi in the right D convention constructs the full localized operation μt(fa,b)=PP⁡u(f(t2+u,t2)Y(a,u)b), u=t1−t2, for the exact left connection ∂t+T and right conversion v dt⋅∂t=−(∂tv+Tv)dt. It proves its actual three-input Beilinson–Drinfeld Jacobi identity, skew-symmetry and polar unit axiom, and identifies its regular restriction with χt. In particular it supplies the local Jacobi premise stated above without an extra assumption on V.

Proof

1.1F1F3F7algebra

Use the full local map of [F7] on every section of the localized two-input module. Its coefficients are finite principal parts for each bounded-weight input and fixed pole order. On regular constant inputs its principal part is precisely the finite sum defining χt: [F1] makes a(j)b=0 for large j, and the definition of PP⁡u gives its coefficients a(j)b directly, with the normalized delta notation fixed by [F3]. No identification of states with operator-valued commutator coefficients is made. On polar inputs one uses the full formula in [F7], including negative vertex products; singular coefficients alone are not substituted for a map on the localization. For a holomorphic finite-pole input, only finitely many normal Taylor coefficients contribute to each output coefficient, by the truncation and weight bounds in [F7]. Those coefficients are holomorphic along the diagonal. Thus these are actual maps on the meromorphic sheaves specified in the Statement, not merely maps on formal completions.

1.2F1F2F7algebra

The right-D convention in [F7] has negative formal adjoint of ∂t+T, with T=L(−1) by [F2]. Its skew-symmetry includes the canonical-bundle and normal-coordinate reversal signs. Its unit is the horizontal map ωX→VXr, fdt↦fdt⊗1: T1=0 by [F1], and μt((t1−t2)−11,b)=bδ(t1,t2) is the canonical residue quotient, not merely vanishing on regular unit inputs. Consequently both the full skew and unit axioms hold in each coordinate.

1.3F2F4F6F7algebra

On an overlap write t′=ρ(t) and put ρy(u)=ρ(y+u)−ρ(y). Apply [F4] to the adjoint field, coefficientwise over the base ring in y. Its coefficients depend on finite jets on bounded weight stages, so the same identity is valid with these formal base coefficients. Since (ρy)u=ρy+u, it gives the complete OPE covariance formula RV(ρy)Y(a,u)b=Y(RV(ρy+u)a,ρ(y+u)−ρ(y))RV(ρy)b. Pair the principal-part map of [F7] with any regular normal test function and take the residue in the first point. Under the substitution, its first one-form factor transforms by dt1′=ρ′(y+u)du and its remaining output one-form by dt2′=ρ′(y)dy. The state factors transform by exactly the operators in the displayed full-field identity. Residue invariance [F6], applied coefficientwise, therefore makes every normal-jet pairing agree in the two coordinates, including for polar source sections. These pairings separate finite principal parts, as in [F7]; hence the full local maps agree on overlaps. This uses full field covariance, not uniqueness of the singular commutator as an asserted determination of polar values.

2.1F1F2F4F5F7step 1.3algebra

The connection also agrees on overlaps in the declared sign. Apply the same covariance identity to the vacuum, fixed by RV, and use [F1] to obtain RV(ρy)euTa=e(ρ(y+u)−ρ(y))TRV(ρy+u)a. Differentiation in u at zero gives ∂yRV(ρy)+ρ′(y)TRV(ρy)=RV(ρy)T. This is exactly the transition identity for the left connection ∂t+T; tensoring with the canonical line and taking its negative formal adjoint gives the right action used in [F7]. The finite-jet coefficients of the local maps vary holomorphically in the base coordinate, and the torsor transitions satisfy the cocycle by [F5]. Thus the local right-D maps glue consistently, including on triple overlaps, to the stated global operation.

3.1F7step 1.1step 1.2step 2.1given∎

The full three-input chiral Jacobi identity in [F7] compares exactly the three localized composites, with the right-D signs established there from the all-integer vertex Jacobi residue identity. Since the local maps agree on overlaps by step 2.1, their composites agree on overlaps as well, and the local identity descends to the global operation. Step 1.2 supplies its skew-symmetry and the actual residue-unit axiom. This establishes every claimed global property with the local Jacobi identity proved by the separate supplier, retaining the Statement's full source, target and convention.

Depends on

Dependency tree · 2 levels

Sources