Alphabeta Math
ExampleConstruction: AI-generatedVerification: AI-generatedPipeline-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 source–sink move in A3: transporting the Euler and skew forms by an initial letter

Statement

Let (W,S) be of type A3, c=s1s2s3 with initial letter s=s1, and let c′=scs=s2s3s1, a reduced Coxeter word for the conjugate Coxeter element s1cs1; in c′ the generator s1 is final instead of initial, a source-sink move. For the forms of c′ one computes, in the ordered basis (es2,es3,es1), Ec′=(100−110−101),ωc′=Ec′−Ec′T=(011−100−100), that is, in the fixed basis (es1,es2,es3) one has Ec′(es1,es2)=−1, Ec′(es1,es3)=0, Ec′(es2,es3)=0 and ωc′(es1,es2)=−1, ωc′(es1,es3)=0, ωc′(es2,es3)=1. Then:

(i) The conjugation identities of The greedy scan computes the c-sorting word; commutation, conjugation and rank-two alignment (3) hold: Ec′(ρ(s1)β,ρ(s1)β′)=Ec(β,β′) and ωc′(ρ(s1)β,ρ(s1)β′)=ωc(β,β′) for all β,β′ taken from the simple basis. For example, using ρ(s1)es1=−es1, ρ(s1)es2=es1+es2 and ρ(s1)es3=es3: Ec′(ρ(s1)es1,ρ(s1)es2)=Ec′(−es1,es1+es2)=−Ec′(es1,es1)−Ec′(es1,es2)=−1+1=0=Ec(es1,es2), ωc′(ρ(s1)es1,ρ(s1)es2)=ωc′(−es1,es1+es2)=−ωc′(es1,es1)−ωc′(es1,es2)=0+1=1=ωc(es1,es2), and similarly for the remaining pairs.

(ii) The orientation of the rank-two subsystem W{s1,s2} changes sign when expressed in its canonical generators: in c the relative order is s1 before s2, in c′ it is s2 before s1, and correspondingly ωc(es1,es2)=1 while ωc′(es2,es1)=1; the transported form is the same form read after applying the reflection ρ(s1) to the two canonical roots. This is the sign convention used in The greedy scan computes the c-sorting word; commutation, conjugation and rank-two alignment (4): the forms transport by conjugation while the orientation on the fixed canonical roots reverses. No Axiom of Choice is used.

Facts & Assumptions

Given: the type-A3 Coxeter matrix with S={s1,s2,s3}, the presented group W, the space V=RS with its simple basis (es1,es2,es3), the Coxeter form B, the canonical reflection representation ρ, the words c=s1s2s3 and c′=s1cs1=s2s3s1, and the forms K=2B, Ec,ωc,Ec′,ωc′.

[F1]

Coxeter elements, the oriented Euler form, the skew form, and the periodic word (1),(2): a Coxeter word uses each element of S once; for a chosen ordered word, K=2B and Ec(esi,esj)=K(esi,esj) when i>j, 1 when i=j, 0 when i<j, with ωc=Ec−EcT.

[F2]

Coxeter words are commutation-connected; the Euler and skew forms depend only on the Coxeter element (1): every Coxeter word is reduced, and every reduced expression of a Coxeter element is again a Coxeter word.

[F3]

The real Coxeter form, its radical, reflections, and form-preserving maps (2),(3): B(es,es)=1, B(es1,es2)=B(es2,es3)=−12, B(es1,es3)=0, and ra(v)=v−2B(v,a)B(a,a)a for B(a,a)≠0.

[F4]

The canonical reflection homomorphism, roots, reflections, and the positive cone (1): ρ:W→GL(V) is a homomorphism with ρ(s)=res for every s∈S.

[F5]

Coxeter elements, the oriented Euler form, the skew form, and the periodic word (2) combined with The real Coxeter form, its radical, reflections, and form-preserving maps (2),(3): for the word c=s1s2s3 the triangular rule with K=2B, B(es1,es2)=B(es2,es3)=−12 and B(es1,es3)=0 gives the simple-basis entries Ec(es1,es2)=0, Ec(es1,es3)=0, Ec(es2,es1)=−1, Ec(es2,es3)=0, Ec(es3,es1)=0, Ec(es3,es2)=−1 and the diagonal entries 1, hence ωc(es1,es2)=Ec(es1,es2)−Ec(es2,es1)=1, ωc(es2,es3)=1 and ωc(es1,es3)=0.

[F6]

The greedy scan computes the c-sorting word; commutation, conjugation and rank-two alignment (3): if s is initial in c, then for all β,β′∈V one has Escs(ρ(s)β,ρ(s)β′)=Ec(β,β′) and ωscs(ρ(s)β,ρ(s)β′)=ωc(β,β′).

[F7]

Proof

technique · identify the conjugate word, compute its forms from the triangular rule, then check the conjugation identity on the nine simple-basis pairs directly
1.1F1F2F7given

The word c′=s1cs1 evaluates to s1s1s2s3s1=s2s3s1 using s12=1 [F7], and it uses each of s1,s2,s3 once, so c′ is a Coxeter word and, by [F2], a reduced Coxeter word for the conjugate Coxeter element s1cs1; in it s1 is the final letter. This is the source-sink move.

1.2F1F3algebragiven

Compute Ec′ in the ordered basis (q1,q2,q3)=(es2,es3,es1) from [F1] and the entries K=2B of [F3]: K(es2,es3)=K(es3,es2)=−1, K(es1,es2)=K(es2,es1)=−1, K(es1,es3)=K(es3,es1)=0. Since s2 precedes s3, which precedes s1 in the word c′, the triangular rule gives Ec′(q1,q1)=Ec′(q2,q2)=Ec′(q3,q3)=1, Ec′(q2,q1)=K(es3,es2)=−1, Ec′(q3,q1)=K(es1,es2)=−1, Ec′(q3,q2)=K(es1,es3)=0, and the three upper-triangular entries Ec′(q1,q2)=Ec′(q1,q3)=Ec′(q2,q3)=0. This is the displayed matrix; subtracting its transpose gives the displayed ωc′. In the fixed basis (es1,es2,es3) the read-off entries are Ec′(es1,es2)=Ec′(q3,q1)=−1, Ec′(es1,es3)=Ec′(q3,q2)=0, Ec′(es2,es3)=Ec′(q1,q2)=0, while ωc′(es1,es2)=Ec′(es1,es2)−Ec′(es2,es1)=−1−0=−1, ωc′(es1,es3)=Ec′(es1,es3)−Ec′(es3,es1)=0, and ωc′(es2,es3)=Ec′(es2,es3)−Ec′(es3,es2)=0−(−1)=1.

1.3F3F4algebra

The reflection ρ(s1)=res1 acts on the simple basis by [F3] and [F4]: ρ(s1)es1=−es1, ρ(s1)es2=es2−2B(es2,es1)es1=es1+es2, and ρ(s1)es3=es3−2B(es3,es1)es1=es3.

2.1step 1.2step 1.3F1F3F5algebra

Check Ec′(ρ(s1)eq,ρ(s1)eq′)=Ec(eq,eq′) on the nine simple pairs, using bilinearity, step 1.3, the entries of step 1.2 and the values of [F5]. Pair (q,q′)=(s1,s2): Ec′(−es1,es1+es2)=−Ec′(es1,es1)−Ec′(es1,es2)=−1+1=0=Ec(es1,es2). Pair (s1,s1): Ec′(−es1,−es1)=Ec′(es1,es1)=1=Ec(es1,es1). Pair (s1,s3): Ec′(−es1,es3)=−Ec′(es1,es3)=0=Ec(es1,es3). Pair (s2,s1): Ec′(es1+es2,−es1)=−Ec′(es1,es1)−Ec′(es2,es1)=−1−0=−1=Ec(es2,es1). Pair (s2,s2): Ec′(es1+es2,es1+es2)=Ec′(es1,es1)+Ec′(es1,es2)+Ec′(es2,es1)+Ec′(es2,es2)=1−1+0+1=1=Ec(es2,es2). Pair (s2,s3): Ec′(es1+es2,es3)=Ec′(es1,es3)+Ec′(es2,es3)=0+0=0=Ec(es2,es3). Pair (s3,s1): Ec′(es3,−es1)=−Ec′(es3,es1)=0=Ec(es3,es1). Pair (s3,s2): Ec′(es3,es1+es2)=Ec′(es3,es1)+Ec′(es3,es2)=0−1=−1=Ec(es3,es2). Pair (s3,s3): Ec′(es3,es3)=1=Ec(es3,es3). All nine pairs agree, so by bilinearity the identity holds on all of V.

3.1step 2.1F1F6algebra

Transport of ω: since ωc′=Ec′−Ec′T and ωc=Ec−EcT by [F1], and since ρ(s1) is linear and preserves the pairing of arguments' roles, step 2.1 gives ωc′(ρ(s1)β,ρ(s1)β′)=Ec′(ρ(s1)β,ρ(s1)β′)−Ec′(ρ(s1)β′,ρ(s1)β)=Ec(β,β′)−Ec(β′,β)=ωc(β,β′) for all basis vectors, hence for all vectors by bilinearity. This proves (i) by direct computation, illustrating the general identity [F6].

4.1F5F6step 1.2step 3.1givenalgebra

Clause (ii): from [F5], ωc(es1,es2)=1, and step 1.2 gives ωc′(es2,es1)=1. In c=s1s2s3 the noncommuting pair s1,s2 occurs in the order s1 before s2, so the oriented entry is ωc(es1,es2)=−K(es1,es2)=1; in c′=s2s3s1 the same pair occurs in the reversed order s2 before s1, and the oriented entry is ωc′(es2,es1)=1, positive in the reversed order. Moreover step 3.1 with β=es1,β′=es2 exhibits the transported identity ωc′(ρ(s1)es1,ρ(s1)es2)=ωc(es1,es2): the new form is the old form read after applying ρ(s1) to the two canonical roots, so the forms are conjugate, while the sign on the fixed ordered canonical roots is reversed. This is the sign convention used in the rank-two alignment definition.

5.1step 1.1step 1.2step 1.3step 2.1step 3.1step 4.1∎

Conclusion: steps 1.1-1.3, 2.1, 3.1 and 4.1 verify (i) and (ii) by exact matrix and basis-pair computations. Every witness is a fixed basis vector or a fixed word, so no Choice is used.

Depends on

Used by

Nothing in the library uses this result yet.

Dependency tree · two levels

54 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