LemmaStatement: Literature-sourcedProof: AI-adaptedPipeline-generatedprecheck passjudge pass (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 q=1 toral specialization identifies the classical Cartan generators

Statement

Let $(I, A, D, P, P^{\lor}, q)$ be a finite symmetrizable Cartan datum, let $U_{A}$ be the integral form with toral part $U_{A}^{0}$ (The Lusztig divided-power integral form $U_{A}$ ) and let $k$ be a characteristic-zero field regarded as an $A$ -algebra by $q \mapsto 1$ . Write $H_{i} : = T_{i, 0, 1}$ and $\overset{ˉ}{x}$ for the image of $x$ in a specialization.

(a) Algebra level. The positive toral quotient $U_{A}^{0} \otimes_{A} k / (K_{h} - 1 : h \in P^{\lor})$ is the polynomial algebra $k [H_{i} : i \in I]$ , and in it ${\overset{ˉ}{T}}_{i, c, t} = (\binom{H_{i} + c}{t})$ for all $c \in Z$ , $t \geq 0$ (Rational-local quantum halves and torus specialize at q=1(c)). No residual $K_{h}$ -ambiguity remains after this quotient — the ideal $(K_{h} - 1 : h \in P^{\lor})$ is exactly the kernel of the comparison with $k [H_{i}]$ — but it is not claimed that $K_{h} = 1$ in the unquotiented specialization $U_{A}^{0} \otimes_{A} k$ .

(b) Label-graded lattices. Let $M$ be a $P$ -graded type-one $U_{q} (g)$ -module with label grading $M = ⨁_{μ \in P} M_{μ}$ (Quantum weight modules, weights in $P$ , and integrability), and let $M_{A} = ⨁_{μ \in P} (M_{A} \cap M_{μ})$ be a $U_{A}$ -stable $A$ -submodule that is graded in the lattice's own sense, i.e. it is the direct sum of its intersections with the label spaces. Then, on $M_{A} \otimes_{A} k$ , the operators induced by $E_{i}, F_{i}$ (and all of $U_{A}$ ) satisfy: ${\overset{ˉ}{K}}_{h} = i d$ for every $h \in P^{\lor}$ ; ${\overset{ˉ}{T}}_{i, c, t}$ acts on the image of $M_{A} \cap M_{μ}$ by the integer $(\binom{⟨ μ, h_{i} ⟩ + c}{t})$ ; and $[{\overset{ˉ}{E}}_{i}, {\overset{ˉ}{F}}_{j}] = δ_{i j} {\overset{ˉ}{H}}_{i}$ , $[{\overset{ˉ}{H}}_{i}, {\overset{ˉ}{E}}_{j}] = a_{i j} {\overset{ˉ}{E}}_{j}$ , $[{\overset{ˉ}{H}}_{i}, {\overset{ˉ}{F}}_{j}] = - a_{i j} {\overset{ˉ}{F}}_{j}$ , where ${\overset{ˉ}{H}}_{i}$ acts on the image of $M_{A} \cap M_{μ}$ by the integer $⟨ μ, h_{i} ⟩$ .

The grading hypothesis in (b) is a real hypothesis on the lattice: it is not inferred from the mere fact that $M_{A}$ spans a weight module. All divisions that occur are performed inside the DVR $R = Q [q]_{(q - 1)}$ (units $[r]_{i}!$ and $q_{i}^{\pm 1}$ ); no vanishing denominator $q_{i}^{s} - q_{i}^{- s}$ is substituted by $q = 1$ .

Facts & Assumptions

Given: The integral form $U_{A}$ , a label-graded type-one module $M$ , a $U_{A}$ -stable graded lattice $M_{A}$ , and the specialization $A \to k$ , $q \mapsto 1$ .

[F1]

The positive toral quotient is $k [H_{i} : i \in I]$ with ${\overset{ˉ}{T}}_{i, c, t} = (\binom{H_{i} + c}{t})$ ; the localized toral $R$ -basis ${K_{c} \prod_{i} K_{i}^{δ_{i}} H_{i}^{t_{i}}}$ exists, and $T_{c, 1} = K_{i} [c]_{i} + q_{i}^{- c} T_{i, 0, 1}$ ; the kernel statement is the equality $U_{A}^{0} \otimes_{A} k / (K_{h} - 1) ≅ k [H_{i}]$ of the cited lemma (Rational-local quantum halves and torus specialize at q=1(b),(c)).

[F2]

On $M_{μ}$ , $K_{h}$ acts by $q^{⟨ μ, h ⟩}$ and $T_{i, c, t}$ acts by the Gaussian binomial ${(\binom{⟨ μ, h_{i} ⟩ + c}{t})}_{i}$ ; the mixed and shifted relations $E_{i} F_{j} - F_{j} E_{i} = δ_{i j} T_{i, 0, 1}$ and $T_{i, 0, 1} E_{j} = E_{j} T_{i, a_{i j}, 1}$ , $T_{i, 0, 1} F_{j} = F_{j} T_{i, - a_{i j}, 1}$ hold as operators on $M$ , and $E_{i}, F_{i}, K_{h}^{\pm 1}, T_{i, c, t}$ preserve $M_{A}$ (Quantum weight modules, weights in $P$ , and integrability, Divided-power multiplication, toral binomials and the $E_{i}^{(a)} F_{i}^{(b)}$ reordering identity, The Lusztig divided-power integral form $U_{A}$ ).

[F3]

For an integer $M \in Z$ , ${(\binom{M}{t})}_{i} = \prod_{s = 1}^{t} [M - s + 1]_{i} / [s]_{i}$ is a Laurent polynomial in $q_{i}$ : for $M \geq 0$ use The quantum Pascal recurrences, the Gauss product formula and Gaussian integrality, and for $M = - a < 0$ the product equals $(- 1)^{t} {(\binom{a + t - 1}{t})}_{i}$ ; its value at $q_{i} = 1$ is the generalized binomial coefficient $(\binom{M}{t}) = \prod_{s = 1}^{t} (M - s + 1) / s$ , an integer, including when $M < 0$ (Quantum integers, factorials, Gaussian binomials and divided powers at $q_{i}$ , Quantum weight modules, weights in $P$ , and integrability). Likewise $q^{⟨ μ, h ⟩} ∣_{q = 1} = 1$ , and the labels $⟨ μ, h_{i} ⟩$ are integers (Kac moody integral and dominant integral weights, Integrable kac moody module).

[F4]

If a graded lattice is the direct sum of its label pieces, then an $A$ -linear operator acts on $M_{A} \otimes_{A} k$ through its values on the pieces: $K_{h}$ acts by the scalars $q^{⟨ μ, h ⟩}$ on the pieces, which specialize to $1$ ; the scalar operators required in (b) are therefore well-defined on the whole specialization. Quantum weight modules, weights in $P$ , and integrability is used for the grading convention.

Verification

technique · quote the algebra-level positive quotient from the rational-local theorem, evaluate the diagonal scalars on the graded lattice at q=1, and reduce the integral operator identities modulo q-1

1.1F1algebra

The algebra-level statement. Part (a) is the toral clause of Rational-local quantum halves and torus specialize at q=1(c): the localized basis ${K_{c} \prod_{i} K_{i}^{δ_{i}} H_{i}^{t_{i}}}$ of $U_{A}^{0}$ specializes and is then quotiented by all $K_{h} - 1$ , leaving the freely generated polynomial algebra $k [H_{i}]$ , and the integral recursion $T_{c + 1, t} = z^{t} T_{c, t} + K^{- 1} z^{- c + t - 1} T_{c, t - 1}$ with $T_{c, 1} = K_{i} [c]_{i} + q_{i}^{- c} T_{i, 0, 1}$ specializes to the Pascal recursion with value $(\binom{H_{i} + c}{t})$ ; the kernel of $U_{A}^{0} \otimes_{A} k ↠ k [H_{i}]$ is exactly $(K_{h} - 1 : h \in P^{\lor})$ . Nothing here is a statement about the unquotiented specialization.

1.2F2F3F4algebra

Scalar operators on a label-graded lattice. Let $μ \in P$ and $x \in M_{A} \cap M_{μ}$ . By [F2], $K_{h} x = q^{⟨ μ, h ⟩} x$ and $T_{i, c, t} x = {(\binom{⟨ μ, h_{i} ⟩ + c}{t})}_{i} x$ . Here $⟨ μ, h_{i} ⟩ \in Z$ by the integrality of the labels [F3], so the binomial is a Laurent polynomial in $q_{i}$ , and after tensoring with $k$ its value is the integer $(\binom{⟨ μ, h_{i} ⟩ + c}{t})$ of [F3]; the scalar $q^{⟨ μ, h ⟩}$ specializes to $1$ . Since $M_{A} = ⨁_{μ} (M_{A} \cap M_{μ})$ by the grading hypothesis, these piecewise scalars define honest operators on all of $M_{A} \otimes_{A} k$ : ${\overset{ˉ}{K}}_{h} = i d$ and ${\overset{ˉ}{T}}_{i, c, t}$ acting by $(\binom{⟨ μ, h_{i} ⟩ + c}{t})$ on the image of each piece. In particular these scalar operators are integer-valued, and ${\overset{ˉ}{H}}_{i}$ acts by $⟨ μ, h_{i} ⟩$ . If the lattice is not graded in this sense, a general lattice element may have weight components outside $M_{A}$ and the scalar operators need not descend; this is the caveat of the statement.

2.1F1F2step 1.2algebra∎

The classical Cartan commutators. All generators $E_{i}, F_{j}, H_{i}, K_{h}$ preserve $M_{A}$ by [F2], so the integral operator identities descend to $M_{A} \otimes_{A} k$ . The mixed relation gives ${\overset{ˉ}{E}}_{i} {\overset{ˉ}{F}}_{j} - {\overset{ˉ}{F}}_{j} {\overset{ˉ}{E}}_{i} = δ_{i j} {\overset{ˉ}{T}}_{i, 0, 1} = δ_{i j} {\overset{ˉ}{H}}_{i}$ ; the shifted toral relations give ${\overset{ˉ}{H}}_{i} {\overset{ˉ}{E}}_{j} = {\overset{ˉ}{E}}_{j} (\binom{{\overset{ˉ}{H}}_{i} + a_{i j}}{1}) = {\overset{ˉ}{E}}_{j} ({\overset{ˉ}{H}}_{i} + a_{i j})$ and ${\overset{ˉ}{H}}_{i} {\overset{ˉ}{F}}_{j} = {\overset{ˉ}{F}}_{j} ({\overset{ˉ}{H}}_{i} - a_{i j})$ , because on the image of $M_{A} \cap M_{μ}$ both sides act by the scalars $⟨ μ, h_{i} ⟩$ resp. $⟨ μ, h_{i} ⟩ + a_{i j}$ and $T_{i, a_{i j}, 1}$ specializes to $(\binom{⟨ μ, h_{i} ⟩ + a_{i j}}{1}) = ⟨ μ, h_{i} ⟩ + a_{i j}$ by step 1.2. Hence the specialized operators satisfy the classical Cartan commutators of (b). No step divides by $q_{i}^{s} - q_{i}^{- s}$ after specializing. The factors $[r]_{i}!$ and $q_{i}^{\pm 1}$ are units in the local DVR; quantum-integer and binomial ratios are evaluated through their Laurent-polynomial identities, not by treating the vanishing $q_{i} - q_{i}^{- 1}$ as a unit.

Depends on

Used by

Direct q=1 substitution in quantum denominators Counterexample
Divided powers and the Lusztig form in quantized sl2 Example
Specializing a quantum highest weight lattice at q=1 Example
Specialization at q=1 recovers the classical enveloping relations Proposition
Specialization of the global bases and the root-of-unity boundary Proposition
Finitely generated free integral weight lattices and deformation of the classical weight character Theorem

Dependency tree · two levels

27 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

Math 263C Quantum Groups (Stanford, 2022), lecture slides 'Lusztig's quantum group. Tilting modules. The Fusion Category in Action' (standard reference, not scraped)
Kyeonghoon Jeong, Seok-Jin Kang and Masaki Kashiwara, Crystal Bases for Quantum Generalized Kac-Moody Algebras, arXiv:math/0305390 (standard reference, not scraped)
George Lusztig, Introduction to Quantum Groups (Progress in Mathematics 110, Birkhauser, 1993; second printing 1994; author-hosted scan) (standard reference, not scraped)