Branching of Weyl-module characters on the diagonal torus #
The character of the Weyl module of a shape μ on GL (n + 1) restricts to the diagonal
torus (kˣ)ⁿ × kˣ as
χ_μ(diag(t, z)) = ∑_ν z ^ (|μ| - |ν|) * χ_ν(diag t),
where ν ranges over the shapes interlacing μ with at most n rows. Setting z = 1
gives the character identity for the upper-left block inclusion GL n → GL (n + 1).
Each interlacing shape occurs once. The formula includes the empty rank and shapes with
more than n + 1 rows, whose Weyl modules vanish.
These are identities of group characters evaluated on the diagonal torus, not isomorphisms of restricted representations. They provide the character calculation for multiplicity-free branching; identifying the representation summands is a separate step.
The proofs combine TauCeti.char_weylRepOfShape_diagonal_eq_eval_diagramSchurPoly, the Weyl-module
character formula, with TauCeti.eval_snoc_diagramSchurPoly, the Schur-polynomial branching
identity. The FDRep forms also apply to TauCeti.schurFunctor over ℂ.
References #
- W. Fulton and J. Harris, Representation Theory: A First Course (1991), Lecture 15 and Appendix A: Schur characters and branching of general linear representations.
On diagonal matrices diag(t, z) of GL (n + 1), the Weyl character branches over
interlacing shapes.
The last coordinate records the number of boxes removed from the shape.
The diagonal-torus branching identity for the bundled Weyl module.
On the diagonal torus, restriction along the upper-left block inclusion branches the Weyl character as a sum over interlacing shapes, each occurring once.
The block-restriction character identity on the diagonal torus for the bundled Weyl
module, equivalently the Schur functor over ℂ.