Alphabeta Math
DefinitionDefinition: Literature-sourcedProof: Not applicablePipeline-generatedaudited 2026-09-27
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 rank-two longest type-A Soergel bimodule

Definition

The parabolic. Keep R=Q[x1,…,xn] with deg⁡xi=2, the simple reflections si=(i i+1) and the invariants Rsi of The standard type-A reflection realization and its polynomial ring, and let 1≤i≤n−2. Put Wi,i+1:=⟨si,si+1⟩≤Sn,RWi,i+1:=Rsi∩Rsi+1, the parabolic subgroup generated by the two adjacent simple reflections and its invariant ring. The two generators act as adjacent transpositions on the three indicated coordinates and fix every other coordinate, so their restriction is an isomorphism Wi,i+1≅S3 by the generation clause of Type-A reduced words and the Coxeter presentation; the subgroup permutes the three coordinates xi,xi+1,xi+2 and fixes every other coordinate, so RWi,i+1 is the ring of polynomials in xi,xi+1,xi+2 that are invariant under all permutations of the three, with all other xj adjoined as invariants. The element wi,i+1:=sisi+1si=si+1sisi+1∈Wi,i+1 is the longest element of Wi,i+1, of length 3.

The bimodule. Define Bi,i+1,i:=R⊗RWi,i+1R(3), the balanced tensor product of the (R,RWi,i+1)-bimodule R with the (RWi,i+1,R)-bimodule R, with the total grading of the tensor product shifted by the external shift (3) of The Soergel bimodule Bi of a simple reflection, that is (3)={−3},(Bi,i+1,i)d=(R⊗RWi,i+1R)d+3  in the convention M(r)d=Md+r. It is the rank-two longest type-A Soergel bimodule attached to i. Equivalently Bi,i+1,i=R⊗RWi,i+1R{−3} in the internal notation M{r}d=Md−r of the library, and when n=3 and i=1 it is B1,2,1=R⊗RS3R(3). The Bott–Samelson bimodule of the word (i,i+1,i) is Bi,i+1,i‾=Bi⊗RBi+1⊗RBi=R⊗RsiR⊗Rsi+1R⊗RsiR(3) in the notation of The Bott–Samelson bimodule of a word.

Remark

(a) Why "longest". For adjacent i,i+1 the rank-two parabolic is finite of type A2, so by the description of the 2mst-valent vertex of Elias–Williamson (p. 6) the indecomposable bimodule Bwi,i+1 indexed by the longest element of the parabolic occurs as a direct summand with multiplicity one in each of the two Bott–Samelson bimodules BiBi+1Bi and Bi+1BiBi+1; the rank-two decomposition theorem of this page identifies that summand explicitly as Bi,i+1,i, so that BiBi+1Bi≅Bi,i+1,i⊕Bi. The reader should not read this remark as a proof: the summand identification is the content of the next item on this page.

(b) The underlying rank. The ring R is free as an RWi,i+1-module of rank ∣Wi,i+1∣=6; this is the classical Chevalley theorem for the rank-two parabolic, and in the case n=3 it is the isomorphism R≅⨁w∈S3RS3(−2ℓ(w)) of RS3-modules used in the source. Consequently Bi,i+1,i, whose grading is shifted by (3), is a free graded R-module of rank six on each side, with homogeneous basis degrees 2ℓ(w)−3 in the convention (r), w∈S3; as a graded left R-module Bi,i+1,i≅R(−3)⊕R(−1)⊕2⊕R(1)⊕2⊕R(3).

(c) The rank-one analogue. For n=2 the parabolic of the single simple reflection is W1=⟨s1⟩ with longest element s1 of length 1. Its analogous invariant-ring formula is BW1:=R⊗Rs1R(1)=B1 of The Soergel bimodule Bi of a simple reflection. This is the rank-one generator, not an instance of the rank-two notation Bi,i+1,i, since two adjacent simple reflections do not exist for n=2.

(d) Degenerate indices. For n≤2 there is no pair of adjacent simple reflections, Wi,i+1 is undefined, and the notation Bi,i+1,i is not used; the objects of the type-A Soergel categories of this page are then generated by the single B1 (or by R alone when n≤1).

(e) Imported facts of Libedinsky used by the decomposition theorem. The rank-two decomposition theorem of this page imports the computation of Libedinsky's §4.4 for the triple product BsBtBs of the rank-two parabolic generated by s≠t; in the type-A notation of this item, with s=si, t=si+1 and the three coordinates xi,xi+1,xi+2, its content is the following. For the idempotent e constructed from that computation, the image im⁡(1−e) is generated as an R-bimodule by the 1-tensor 1⊗:=1⊗1⊗1⊗1∈Bi,i+1,i‾, and BiBi+1Bi is generated as an R-bimodule by 1⊗ together with 1⊗xi⊗1⊗1. Moreover there is a graded homomorphism of R-bimodules R⊗RWi,i+1R(3)⟶Bi,i+1,i‾,p⊗q⟼p⊗1⊗1⊗q, whose image is the sub-bimodule ⟨1⊗⟩ generated by the 1-tensor; since both sides are isomorphic to R(−3)⊕R(−1)⊕2⊕R(1)⊕2⊕R(3) as graded left R-modules, this homomorphism is an isomorphism onto im⁡(1−e). These four statements are quoted from Libedinsky, §4.4.1 and §4.4.2, and are used in the decomposition theorem of this page, not proved here.

Depends on

Used by

Dependency tree · two levels

11 results within two dependency steps of this one, each drawn at its shortest distance from it. An arrow runs from a result to what uses it, so the chart reads left to right and ends at this result, which carries a heavier outline. Every node is a link to that result. Click elsewhere on the chart to enlarge it.

Sources