Alphabeta Math
ExampleConstruction: Literature-sourcedVerification: AI-adaptedPipeline-generatedprecheck passaudited 2026-10-08
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 point outside the Tits cone with infinite stabilizer

Example

Let S={s1,s2,s3,s4} with m(s1,s2)=m(s3,s4)=∞, m(si,sj)=2 whenever one of si,sj lies in {s1,s2} and the other in {s3,s4}, and m(si,si)=1; thus W=W1×W2 with W1=⟨s1,s2⟩ and W2=⟨s3,s4⟩ two infinite dihedral groups (Coxeter matrices, the presented Coxeter group, reduced words, length, and standard parabolic subgroups). Write f=(x1,x2,x3,x4) for a functional, put Δ1:=x1+x2 and Δ2:=x3+x4, and let U be the Tits cone of (W,S) (The Tits cone, its interior, and the negative-root set of a functional). Then:

(i) Block decomposition. B is block diagonal with blocks (1−1−11) on {es1,es2} and on {es3,es4} and zero cross terms, each generator acts nontrivially only on its own block, the dual action is componentwise, and C=C1×C2,U=U1×U2, where Ci, Ui are the chamber and the Tits cone of the i-th infinite dihedral factor.

(ii) The vector. By The Tits cone of infinite dihedral type: interior, boundary, and stabilizers (iii), U1={f1:Δ1(f1)>0}∪{0}. The functional f=(−1, 0, 1, −1) has Δ1(f)=−1<0, hence f1∉U1 and f∉U.

(iii) Infinite stabilizer. Stab⁡W(f)={1,s2}×⟨s3s4⟩: the first factor is the order-two subgroup fixing the line x2=0 pointwise, the second is infinite cyclic. Hence f is a vector outside the Tits cone whose stabilizer is infinite, so it is not a finite parabolic subgroup of W; this shows that an arbitrary vector outside the Tits cone need not have a finite parabolic stabilizer. For contrast, in the rank-two example The Tits cone of infinite dihedral type: interior, boundary, and stabilizers the infinite stabilizer ⟨st⟩ occurs exactly on the boundary line Δ=0 with the origin removed, which lies outside U but in its closure, and every point with Δ≠0 has stabilizer of order at most 2: no uk with k≠0 fixes it, and uks⋅f=f reduces to the single equation xs=−kΔ, which determines at most one k. Here moreover Δ1(f1)=−1<0, so f also lies outside the closure U‾=U1‾×U2‾={f:Δ1(f)≥0, Δ2(f)≥0} of the Tits cone, and the infinite stabilizer is carried by a point strictly outside the closed cone.

Facts & Assumptions

Given: S={s1,s2,s3,s4} with m(s1,s2)=m(s3,s4)=∞, m(si,sj)=2 across the two blocks and m(si,si)=1; the presented group W with W1=⟨s1,s2⟩ and W2=⟨s3,s4⟩; a functional f=(x1,x2,x3,x4) with Δ1:=x1+x2 and Δ2:=x3+x4; the Tits cone U of (W,S) as in The Tits cone, its interior, and the negative-root set of a functional.

[F1]

For any pair (s,t) with m(s,t)=∞ in a Coxeter system, the subgroup W{s,t} is the Coxeter group presented by the restricted rank-two matrix, so the following rank-two facts apply to it: the generators act by s:(xs,xt)↦(−xs, 2xs+xt) and t:(xs,xt)↦(xs+2xt, −xt), and Δ is invariant; with u=st one has uk⋅x=x+2kΔ(x)(−1,1) and uks⋅x=(−(xs+2kΔ(x)), xt+2xs+2kΔ(x)) for all k∈Z, and W={uk}∪{uks}; the Tits cone is U={f:Δ(f)>0}∪{0} with closure U‾={Δ≥0}; and Stab⁡W(f)=⟨st⟩ for every f with Δ(f)=0 and f≠0, while every point of U∖{0} has stabilizer of order at most 2. (The Tits cone of infinite dihedral type: interior, boundary, and stabilizers (i)-(iii), (v), Support, intrinsic parabolic presentations, minimal coset representatives and length additivity, with the type-A identification (2)).

[F2]

U=⋃w∈WwC for the closed chamber C={f:f(es)≥0 for all s}, and wU=U for every w∈W. (The Tits cone, its interior, and the negative-root set of a functional (1)-(2)).

[F3]

For f∈C one has Stab⁡W(f)=WS(f), and f∈U if and only if Neg⁡(f) is finite. (Chamber collisions, point stabilizers, and the intersection rule (4), The finite-negativity criterion, the reduction step, and convexity of the Tits cone (1)).

[F4]

For a cross pair one has B(ei,ej)=−cos⁡(π/2)=0, while B(es1,es2)=B(es3,es4)=−1 and all diagonal entries are 1; for B(a,a)=1 the reflection is rav=v−2B(v,a)a. (The real Coxeter form, its radical, reflections, and form-preserving maps (2)-(3), Quarter-turn values and shifts by pi/2 and pi).

[F5]

Each rs is a linear involution, and for m(s,t)=∞ the product rsrt has infinite order on V. (Reflections: involutivity, form invariance, fixed hyperplane, and exact rank-two order (2), (3)(iv)).

[F6]

ρ(s)=rs for every generator s; the dual action is (w⋅f)(v)=f(ρ(w)−1v); and C={f:f(es)≥0 for all s}. (The canonical reflection homomorphism, roots, reflections, and the positive cone (1), The dual action, chambers, faces, and root hyperplanes (1)-(2)).

[F7]

W is the presented group of the Coxeter matrix: generators are involutions, a relator entry m(s,t)=2 means (st)2=1, and every assignment of the generators to elements of a group satisfying these relations extends uniquely to a homomorphism of W. (Coxeter matrices, the presented Coxeter group, reduced words, length, and standard parabolic subgroups).

Verification

technique · componentwise computation in a product of two infinite dihedral groups
1.1F1F2F4F5F6F7algebra

Block decomposition, (i). For a cross pair the Coxeter entry is 2, so the corresponding off-diagonal form entry is −cos⁡(π/2)=0, and the form is block diagonal with blocks (1−1−11) on {es1,es2} and on {es3,es4}. For a generator s of one block and a basis vector ej of the other block, the reflection formula gives rsej=ej−2B(ej,es)es=ej, so each generator acts trivially on the other block. Put Vi:=span⁡{ej:j∈Si}; the same computation shows that every ρ(s) preserves each Vi and that ρ decomposes as ρ(w1,w2)=ρ1(w1)⊕ρ2(w2), once W is identified with W1×W2: cross generators s∈S1, t∈S2 satisfy (st)2=1, hence st=t−1s−1=ts because both are involutions, so the blocks commute. The assignment s1↦(s1,1), s2↦(s2,1), s3↦(1,s3), s4↦(1,s4) respects the relators (within-block relators hold factorwise and cross pairs satisfy ((si,1)(1,sj))2=(si2,sj2)=1), so by the universal property of [F7] it induces a homomorphism Φ:W→W1×W2, which is inverse to the multiplication map W1×W2→W because both composites are homomorphisms agreeing with the identity on generators. By the intrinsic parabolic presentation each Wi is moreover the Coxeter group presented by the restricted rank-two matrix m∣Si (Support, intrinsic parabolic presentations, minimal coset representatives and length additivity, with the type-A identification (2)), so the rank-two facts of [F1] apply to W1 and W2. The dual action is therefore componentwise, and C={f:f(ej)≥0 for all j}={f1:f1(es1),f1(es2)≥0}×{f2:f2(es3),f2(es4)≥0}=C1×C2, whence U=⋃(w1,w2)w1C1×w2C2=U1×U2. This is (i).

2.1F1step 1.1algebra

The vector, (ii). By clause (iii) of the rank-two example, U1={f1:Δ1(f1)>0}∪{0}. For f=(−1,0,1,−1) one has Δ1(f1)=−1<0, so f1∉U1; by step 1.1 U=U1×U2, hence f∉U.

3.1F1F3F5step 1.1algebra∎

The stabilizer and the contrast, (iii). For the first factor, use the rank-two formulas with (s,t)=(s1,s2) read off from the infinite dihedral example: s1:(x1,x2)↦(−x1, 2x1+x2) and s2:(x1,x2)↦(x1+2x2, −x2), so s2 fixes the line x2=0 pointwise, and in particular fixes f1=(−1,0); writing u1:=s1s2 one has u1k⋅x=x+2kΔ1(x)(−1,1), so u1k⋅f1=(−1+2k,−2k) equals f1 only for k=0, while u1ks1⋅f1=(1+2k,−2−2k) equals (−1,0) only for k=−1, which is the element u1−1s1=s2; since W1={u1k}∪{u1ks1}, this gives Stab⁡W1(f1)={1,s2}. For the second factor f2=(1,−1) has Δ2(f2)=0 and f2≠0, so the rank-two stabilizer computation gives Stab⁡W2(f2)=⟨s3s4⟩, which is infinite cyclic because ρ(s3s4)=rs3rs4 has infinite order. Since the action is componentwise by step 1.1, an element (w1,w2) fixes f exactly when w1 fixes f1 and w2 fixes f2, so Stab⁡W(f)={1,s2}×⟨s3s4⟩; this subgroup is infinite (it contains (1,(s3s4)k) for k≠0), hence it cannot be a finite parabolic subgroup of W. Moreover f lies outside the closure: U⊆{Δ1≥0, Δ2≥0} and this set is closed, so U‾⊆{Δ1≥0, Δ2≥0}, and conversely every point (x1,x2,x3,x4) of it is approached by (x1+ε,x2+ε,x3+ε,x4+ε) as ε>0 tends to zero; these approximants have both Δi>0 and lie in U; since Δ1(f1)=−1<0, the functional f is strictly outside U‾. For contrast, in the rank-two example the infinite stabilizer ⟨st⟩ occurs exactly on the boundary line Δ=0 with the origin removed, which lies outside U but in its closure, and every point with Δ≠0 has stabilizer of order at most 2: no uk with k≠0 fixes it, and uks⋅f=f reduces to the single equation xs=−kΔ, which determines at most one k, the corresponding element uks being an involution. This is (iii).

Depends on

Used by

Nothing in the library uses this result yet.

Dependency tree · two levels

89 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