Alphabeta Math
ExampleConstruction: AI-generatedVerification: AI-generatedPipeline-generatedprecheck passjudge pass (gpt-6-luna)verified 2026-11-04 (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.

The sl2 principal admissible reduction data at level minus one half

Physical scope

A finite exact matrix and root-data computation for sl2 over C. It verifies a principal nilpotent, its chosen good grading and Slodowy slice, and the level k=-1/2 admissibility inequalities; it asserts no BRST cohomology, PBW theorem, simple W quotient, rationality, or empirical claim.

Example

In sl2(C) set e=(0100),h=(100−1),f=(0010). Use the invariant form (X∣Y)=tr⁡(XY), the positive root α with α(h)=2, and the grading element h/2. The chosen triple is (e,h,f), the degree-1 space is Ce, the degree-0 space is Ch, the degree-−1 space is Cf, and g1/2=0. Its Slodowy slice is f+Ce. For this root datum h=h∨=2, r∨=1, and the Langlands-dual root datum is the same. Taking p=3, q=2 gives k=−2+3/2=−1/2, which satisfies the first nondegenerate principal-admissible level test.

Here “principal” means nilpotent with centralizer dimension equal to rank⁡g.

f(h-weight¡2)h(h-weight0)e(h-weight2)adeade

The arrows show the adjoint action of e on the three weight spaces; their coefficients are [e,f]=h and [e,h]=−2e. The picture records the weights and actions only; the verification below is by matrix multiplication.

Facts & Assumptions

[F1]

Finite-dimensional simple complex Lie algebra: simple means nonabelian with no nonzero proper ideals.

[F2]

Nilpotent element in a finite-dimensional Lie algebra: f is nilpotent when ad⁡f is nilpotent.

[F3]

Good grading and isotropic data for affine W-algebra reduction: a good grading has f∈g−1, with ad⁡f injective in degrees j≥1/2 and surjective in degrees j≤1/2; the form ωf on degree 1/2 determines the isotropic datum.

[F4]

Slodowy slice associated to a nilpotent element: for a chosen triple, Sf=f+ge.

[F5]

Root normalization and dual data for principal W levels and Nondegenerate principal admissible affine level: the normalized root data and the two arithmetic level tests have the conventions stated above. The dual datum here is verified directly, so no general dual-realization theorem is used.

Proof

1.1givenF1algebra

Direct multiplication gives [h,e]=2e, [h,f]=−2f, and [e,f]=h. Thus the three matrices are linearly independent and span the traceless 2×2 matrices. The algebra is nonabelian.

2.1F1step 1.1algebra

Let I be an ideal and X=ae+bh+cf∈I. Since I is stable under ad⁡h, it is stable under polynomials in that operator. The eigenvalues on e,h,f are 2,0,−2; the polynomials A(A+2)8,1−A24,A(A−2)8,A=ad⁡h, project onto the e,h,f lines, respectively. If I is nonzero, one of these projections puts a nonzero multiple of one of e,h,f in I. Bracketing that element with the other two basis elements using step 1.1 puts all three in I. Therefore I=sl2, proving simplicity.

2.2F2step 1.1algebra

The brackets are [f,e]=−h, [f,h]=2f, and [f,f]=0, so (ad⁡f)3=0. Hence f is nilpotent by [F2]. The relations in step 1.1 also verify the chosen Jacobson-Morozov triple directly, without invoking its general existence theorem.

2.3F5step 1.1algebra

By step 1.1, ad⁡h is diagonal on the basis e,h,f with eigenvalues 2,0,−2. Its zero eigenspace is Ch, so ad⁡h is diagonalizable and the centralizer of Ch is itself; hence Ch is a Cartan subalgebra of rank 1. The two nonzero weights are α and −α with α(h)=2. For X=ae+bh+cf, [f,X]=−ah+2bf, so gf=Cf and has dimension 1. Thus f is principal under the stated convention.

2.4F3step 1.1algebra

Since ad⁡(h/2) has eigenvalues 1,0,−1 on e,h,f, the displayed grading is a Lie grading. The only nonzero required map with j≥1/2 is ad⁡f:Ce→Ch, e↦−h, which is injective. For j≤1/2, the nonzero target cases are Ch→Cf, h↦2f, which is surjective, and Cf→0, which is also surjective. All other graded pieces are zero. Since g1/2=0, its skew form is nondegenerate on the zero space and the unique Lagrangian is ℓ=0. Thus the grading is good and m=Ce.

2.5F4step 1.1algebra

For X=ae+bh+cf, [e,X]=−2be+ch. This vanishes exactly when b=c=0, so ge=Ce. By [F4], Sf=f+Ce.

3.1F5step 2.3algebra

The trace form has (h∣h)=2 and (e∣f)=1, so α has squared length 2 and α∨=h. Thus θ=α, ρ=α/2, h=1+ht⁡(α)=2, and h∨=1+ρ(h)=2. The coroot root set is again {α,−α}, so its dual datum has hg~∨=2; there are no off-diagonal Cartan entries, giving r∨=1.

4.1F5step 3.1algebra∎

With p=3,q=2, gcd⁡(p,q)=1 and gcd⁡(q,r∨)=gcd⁡(2,1)=1. The first branch requires p≥h∨ and q≥h; both hold because 3≥2 and 2≥2. Thus k=−h∨+p/q=−2+3/2=−1/2, and k+h∨=3/2>0. This verifies the declared level condition and noncriticality.

Proof status

Every claim in this example is checked directly for the explicit rank-one data. It uses the general definitions only at their stated interfaces and makes no claim about the BRST cohomology or W-algebra representation theory.

Depends on

Dependency tree · 2 levels

Sources