Pith. sign in

REVIEW 2 major objections 6 minor 29 references

Algorithm for computing canonical bases and foldings of quantum groups

T0 review · 2 major / 6 minor · reviewed 2026-08-07 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read The paper proves that the transition matrix from the PBW basis to the canonical basis of the negative half of any finite-type quantum group is computed from inner products of monomials, extending the method to non-symmetric types by…

desk verdict Extends Antor's algorithm to non-symmetric finite type with a solid KLR-based inner product formula, but the proof as written ships with a load-bearing omitted E6 computation. read the letter →

arxiv 2506.00793 v1 pith:3UT2FHPN submitted 2025-06-01 math.QA

Toshiaki Shoji , Zhiping Zhou This is my paper
classification math.QA MSC 17B3716G20
keywords canonicalbasisPBWmonomialfoldingquantumgroupsKLRalgebrasMackeyfiltrationtransitionmatrix
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper establishes that the transition matrix between the triangular PBW basis and the canonical basis of the negative half of any finite-type quantum group can be found by a fixed algorithm: compute the inner-product matrix of a suitably chosen monomial basis, then run a unitriangular factorization. This was known for symmetric Cartan data; the paper extends it to the non-symmetric types (B, C, F, G) by folding a symmetric quantum group by an automorphism and constructing a monomial basis compatible with that folding. It proves a closed formula for the required inner products using the KLR-algebra categorification of the folded group and a Mackey filtration, and shows that the folded group's matrix is completely determined by the σ-fixed part of the symmetric group's matrix. A sympathetic reader would care because the canonical basis is central to quantum-group representation theory but is hard to compute directly, and this reduces the computation to elementary linear algebra over monomials.

What carries the argument

The load-bearing object is the monomial basis: a basis of U_q^- whose elements are ordered products of divided powers $f_i^{{(d)}}$ and whose expansion in the PBW basis has the same leading term plus higher-order corrections. The algorithm runs on the matrix identity Λ = ^t H D H relating the monomial inner-product matrix to the diagonal PBW inner-product matrix; because H is lower unitriangular and D diagonal, H and D are recovered from Λ by linear algebra, and the subsequent factorization H = P Q with P lower unitriangular over qZ[q] and Q bar-invariant uniquely gives the PBW-to-canonical transition matrix P. For the non-symmetric case, the paper uses folding: a non-symmetric Cartan datum is realized as the σ-fixed part of a symmetric datum with an admissible automorphism, and a modified monomial basis is built so that σ permutes it. The inner products are computed through a categorification by KLR algebras (graded algebras whose Grothendieck groups realize U_q^-), with a Mackey filtration of induction–restriction functors yielding the sum over permutation matrices with weights $q^{{-A(ξ)}}$.

What would settle it

Evaluate δ(eck) from formula (4.12.1) for the 23 unlisted values in the type E6 case (k = 1, 3, ..., 23); if any of them is nonzero, the asserted monomial basis for the folded type F4 fails to exist and the algorithm of Theorem 6.2 does not apply to F4.

Watch

Extended reading notes

Core claim

On the paper's own terms: for U_q^- of finite type with a fixed reduced expression h, let X_h be the PBW basis, B_h the canonical basis, and M_h a monomial basis whose elements are ordered products of divided powers $f_i^{{(d)}}$. The paper proves that the matrix identity Λ = ^t H D H, with H lower unitriangular over A = Z[q,$q^{{-1}}$] and D diagonal, determines the transition matrix P from X_h to B_h through the unique factorization H = P Q, where off-diagonal entries of P lie in qZ[q] and Q is bar-invariant. For non-symmetric types, it constructs a monomial basis from the symmetric case via folding, proves the needed δ(eck)=0 degree checks in types A2n−1, Dn, D4, G2 and states them for E6, and derives a closed formula for Λ by computing pairings of projective objects in the folded KLR categorification. The comparison theorem then states that Λ for the folded group is obtained from the σ-stable part Λσ of the symmetric group's monomial inner-product matrix by replacing quantum factorials and restricting the permutation sum, so P for the folded group is computed from monomial data for the symmetric group.

Load-bearing premise

The construction of the monomial basis for non-symmetric types rests on the unverified claim that the degree shift δ(eck) is zero in every case; for type E6 the paper states the setup but omits the checks for k = 1, 3, ..., 23, and if any of those is nonzero the F4 monomial basis—and with it Theorem 6.2 for F4—would not follow.

Editorial extensions

If this is right

  • The transition matrix P from PBW to canonical basis is computable for every finite-type quantum group once the monomial basis and its inner products are known, so no separate construction of the canonical basis is needed.
  • For the folded types B_n, C_{n-1}, F_4 and G_2, the inner-product matrix Λ is entirely determined by the σ-fixed part Λσ of the corresponding symmetric type, and P is determined over Z[q,q^{-1}] rather than just modulo p.
  • The same identity explains the known congruence Pσ ≡ P mod p, since the quantum factorials and the exponent sums A(ξ) agree modulo p for σ of prime order.
  • In the trivial-automorphism case the same KLR-based formula gives a closed computation of Λ, so the algorithm also supplies a uniform symmetric-type computation of P.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • A natural next step would be to run the omitted E6 checks explicitly; a single nonzero δ(ec_k) would show the folding-compatible monomial basis needs a different construction for F4, even though the matrix identity itself is unaffected.
  • The same Mackey-filtration computation may carry over to affine or general Kac–Moody foldings, where the coefficient-ring reduction to A is already known, potentially yielding an algorithm for canonical bases beyond finite type.
  • The explicit formula suggests a purely combinatorial reformulation of P: everything reduces to enumerating permutation matrices with a degree weight A(ξ), which could be tabulated independently of the quantum-group machinery.
  • One could implement the algorithm in a computer algebra system for small weights in types B2 and G2 and compare the resulting P with the examples in Section 7; agreement would be a direct consistency check of Theorem 6.7.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

2 major / 6 minor

Summary. The paper extends Antor's algorithm for computing the transition matrix P between the PBW basis and the canonical basis of a quantum group to the non-symmetric finite type case. It constructs monomial bases for U_q^- by folding symmetric types, derives a closed formula for the inner products of monomials using McNamara's KLR categorification with automorphisms (Theorem 5.20), and obtains an algorithm for P (Theorem 6.2). It then compares the unfolded and folded algorithms (Theorem 6.7), with worked examples for B2 and G2.

Significance. If the gaps identified below are filled, this is a valuable contribution: it gives a uniform algorithmic route to canonical bases for all finite types, and it makes the folding relationship between the relevant inner product matrices explicit. The paper has real strengths: Proposition 1.10 gives a clean uniqueness and reconstruction argument; Theorem 5.20 is a concrete closed formula; the B2 and G2 examples are worked in detail and appear consistent. The use of published categorification results (McNamara, Lusztig, MSZ) is appropriate and not circular.

major comments (2)
  1. [§4.13(C), Lemma 4.13] For type E6 the proof of δ(ec_k) = 0 is explicitly omitted: after listing β1,...,β36 and ec1,...,ec24, the text says the required computation is done 'as in the case (D)' and 'we omit the details.' This verification is load-bearing: Lemma 4.15 invokes Lemma 4.13 to identify F(d^k) with b(c^k,h), Theorem 4.16 uses Lemma 4.15 to establish the triangular monomial basis, and Proposition 4.18 uses that basis for the folding-compatible statement; consequently Theorem 6.2 for F4 and the F4 part of Theorem 6.7 rest on it. The gap is finite and probably fillable, but as written the F4 case is unproved.
  2. [§6.8, Remark 6.8] Theorem 6.7 depends on identifying the sequence ν obtained from Pi,d with the expression obtained from Pj,d, up to mutually commuting generators. The remark states that the discrepancy can be ignored because the monomial bases are the same, but no argument is given that the two expressions yield the same element em(c,h), the same sum over Ξ, the same values A(ξ), and the same factors γ. Please replace this remark by a proof or a precise normalization lemma; otherwise the comparison of Λσ and Λ is not established.
minor comments (6)
  1. [§1, Introduction] 'MacKey formula' should be 'Mackey formula'.
  2. [§4.13(D)] The line 'Here N = 6 and N = 12' is ambiguous; with the notation of §3.5 the first N should be the length for D4 and the second the length for G2, but the overlines are missing. Please clarify which N is which.
  3. [§7.2] The four matrices ξ1,...,ξ4 in Ξ(ν,ν′) are printed identically; as displayed, the reader cannot verify the values A(ξ1)=-1, A(ξ2)=1, A(ξ3)=1, A(ξ4)=3. Please show the actual permutation matrices or the corresponding permutations.
  4. [§2.5, proof of Proposition 2.5] 'hods' appears twice and should be 'holds'.
  5. [§4.3(C), display (4.3.8)] The quiver orientation is not legible; please spell out the arrows explicitly.
  6. [§6.2, Theorem 6.2] The theorem states 'quantum group of general type', while the abstract and the body of the paper treat finite type. Please either restrict the statement to finite type or justify the broader wording.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity found: the algorithm derives P from independently computed inner products; self-citations are previously published theorems, and the E6 omission is a proof gap rather than a circular step.

full rationale

The central derivation chain is not circular. Proposition 1.10 (taken from Antor) shows that the monomial inner-product matrix Lambda determines the transition matrices H, D, P, Q through Lambda = tHDH and H = P Q, with uniqueness and an explicit algorithm. The paper's own contribution is to make Lambda computable for non-symmetric types: Theorem 5.20 computes the inner products from McNamara's KLR categorification, an external result cited from [M], not from the unknown matrix P. The folding comparison in Theorem 6.7 derives the folded matrix Lambda from the sigma-fixed matrix Lambda_sigma using the injective map phi of Lemma 6.6; this is a mathematical reduction, not a renaming or a fitted input. The cited results [SZ1], [MSZ2], and [M] are load-bearing but are independent published theorems with their own proofs. In particular, Theorem 3.3 from [SZ1] is a prior result by the same authors, but it is not the target result of this paper and it is not justified by the present algorithm. The only concrete weakness is an omitted verification: in Lemma 4.13(C) the authors write 'We need to show that delta(ec_k)=0 for k=1,3,5,...,21,23. This is done by a similar computation as in the case (D), once the tables for beta_k and ec_k are given. So we omit the details.' This is a genuine proof gap for the E6-to-F4 monomial basis construction, and it is load-bearing for Theorem 6.2 applied to F4, but it is not circular: if one of those computations failed, the result would be a false lemma, not a derivation that assumes its own conclusion. No step in the paper reduces the prediction of P to the definition of P, and no fitted parameter is renamed as a prediction. Therefore no circular step is exhibited, and the circularity score is 0.

Assumptions & free parameters 0 free parameters · 7 assumptions · 0 invented entities

No free parameters or invented entities. All ingredients are imported from published theorems; the main fresh risk is the unproved E6 case in Lemma 4.13 and the identification in Remark 6.8.

assumptions (7)
  • domain assumption Lusztig's canonical basis B exists for U_q^- of finite type and is independent of the reduced expression h, via Saito's crystal basis result.
    Invoked in Remark 1.6 and throughout; the transition matrix P is defined relative to this basis.
  • domain assumption Lusztig's monomial basis theorem [L2, Thm 4.7] for symmetric U_q^-.
    Starting point of Section 4; the folded construction mimics it.
  • domain assumption Folding isomorphism of [SZ1, Thm 0.4] and the PBW/canonical basis bijections of [SZ1, Thm 1.14] (Prop 3.7).
    Bridges the symmetric and non-symmetric algebras; used in Prop 4.18 and Lemma 4.15.
  • domain assumption McNamara's categorification theorem [M, Thm 6.1] and Mackey filtration [M, Thm 4.5] for KLR algebras with automorphisms.
    Provides the isometry gamma and the filtration used in Theorem 5.20.
  • domain assumption Steinberg's expression w0 = c^(h/2) for Coxeter elements and Kirillov's adapted quiver orientation [Ki, 3.33].
    Used in Section 4.3 to choose reduced sequences h and h and orientations for each folding type.
  • standard math The Levendorskii-Soibelman and Xi commutation formulas for root vectors (Prop 2.2, 2.3).
    Used in Section 2 to establish PBW product expansions.
  • standard math The inner product formula (1.4.1) for PBW bases from [L4, 38.2].
    Gives the diagonal matrix D used in the matrix equations.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Algorithm for computing canonical bases and foldings of quantum groups." pith.science (2026). https://pith.science/paper/3UT2FHPN

@misc{pith2026250600793,
  author       = {Pith},
  title        = {Pith review of: Algorithm for computing canonical bases and foldings of quantum groups},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/3UT2FHPN}},
  note         = {Machine review of arXiv:2506.00793}
}
abstract

Let ${\mathbf U}_q^-$ be the negative half of a quantum group of finite type. Let $P$ be the transition matrix between the canonical basis and a PBW basis of ${\mathbf U}_q^-$. In the case ${\mathbf U}_q^-$ is symmetric, Antor gave a simple algorithm of computing $P$ by making use of monomial bases. By the folding theory, ${\mathbf U}_q^-$ (symmetric, with a certain automorphism) is related to a quantum group $\underline{{\mathbf U}}_q^-$ of non-symmetric type. In this paper, we extend the results of Antor to the non-symmetric case, and discuss the relationship between the algorithms for ${\mathbf U}_q^-$ and for $\underline{\mathbf U}_q^-$.

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

29 extracted references · 24 canonical work pages

  1. [1]

    Antor; Canonical bases via pairing monomials, preprint, arXiv:2308.16254v1

    J. Antor; Canonical bases via pairing monomials, preprint, arXiv:2308.16254v1

    arXiv

  2. [2]

    Chari and N

    V. Chari and N. Xi; Monomial bases of quantized enveloping algebras, In ``Recent developments in quantum affine algebras and related topics (Raleigh, NC, 98)'', Contemp. Math. 248 (1999), 69-81

    work page 1999

  3. [3]

    Deng and J

    B. Deng and J. Du; Bases of quantized enveloping algebras, Pacific J. Math. 220 (2005), 33-48

    work page 2005

  4. [4]

    B. Deng, J. Du, B. Parshall and J. Wang; ``Finite dimensional algebras and quantum groups'', Math. Surveys and Monographs, 150 , Amer. Math. Soc. Providence, Rhode Island, 2007

    work page 2007

  5. [5]

    Grojnowski and G

    I. Grojnowski and G. Lusztig; A comparison of bases of quantized enveloping algebras, in ``Linear Algebraic Groups and their Representations'', Contemp. Math. 153 , 1993, pp.11-19

    work page 1993

  6. [6]

    J. E. Humphreys; Reflection groups and Coxeter groups, Cambridge Studies in Adv. Math. 29 , Cambridge Univ. Press, Cambridge, 1990

    work page 1990

  7. [7]

    Kashiwara; On crystal bases of the Q -analogue of universal enveloping algebras, Duke Math

    M. Kashiwara; On crystal bases of the Q -analogue of universal enveloping algebras, Duke Math. J. 63 (1991), 465-516

    work page 1991

  8. [8]

    Kirillov; ``Quiver Representations and Quiver Varieties'', Graduate Studies in Math

    A. Kirillov; ``Quiver Representations and Quiver Varieties'', Graduate Studies in Math. 174 , Amer. Math. Soc. Providence, Rhode Island, 2016

    work page 2016

Show all 29 references
  1. [9]

    Khovanov and A

    M. Khovanov and A. D. Lauda; A diagrammatic approach to categorification of quantum groups, I, Represent. Theory, 13 (2009), 309-347

    2009

  2. [10]

    Khovanov and A

    M. Khovanov and A. D. Lauda; A diagrammatic approach to categorification of quantum groups, II, Transactions of AMS, 363 (2011), 2685-2700

    2011

  3. [11]

    Lusztig; Quantum groups at roots of 1, Geom

    G. Lusztig; Quantum groups at roots of 1, Geom. Dedicata 35 (1990), 89-114

    1990

  4. [12]

    Lusztig; Canonical bases arising from quantized enveloping algebras, J

    G. Lusztig; Canonical bases arising from quantized enveloping algebras, J. Amer. Math. Soc. 3 (1990), 447-498

    1990

  5. [13]

    Lusztig; Quivers, perverse sheaves, and quantized enveloping algebras, J

    G. Lusztig; Quivers, perverse sheaves, and quantized enveloping algebras, J. Amer. Math. Soc. 4 (1991), 365-421

    1991

  6. [14]

    Lusztig; Introduction to quantum groups, Progress in Math

    G. Lusztig; Introduction to quantum groups, Progress in Math. Vol. 110 Birkhauser, Boston/Basel/Berlin, 1993

    1993

  7. [15]

    Lusztig; Piecewise linear parametrization of canonical bases, Pure and Applied Math

    G. Lusztig; Piecewise linear parametrization of canonical bases, Pure and Applied Math. Quart. 7 (2011), 783-796

    2011

  8. [16]

    Levendorskii and Y

    S. Levendorskii and Y. Soibelman; Some applications of quantum Weyl groups, J. Geom. and Physics, 7 (1990), 241-254

    1990

  9. [17]

    P. J. McNamara; Folding KLR algebras, J. London Math. Soc (2), 100 (2019), 447-469

    2019

  10. [18]

    Y. Ma, T. Shoji and Z. Zhou; Diagram automorphisms and canonical bases for quantized enveloping algebras, J. Algebra 614 (2023), 712-753

    2023

  11. [19]

    Y. Ma. T. Shoji and Z. Zhou; Foldings of KLR algebras, J. Algebra 639 (2024), 60-98

    2024

  12. [20]

    Reineke; Feigin's maps and monomial bases for quantized enveloping algebras, Math

    M. Reineke; Feigin's maps and monomial bases for quantized enveloping algebras, Math. Z. 237 (2001), 639-667

    2001

  13. [21]

    Reineke; Quivers, desingularizations and canonical

    M. Reineke; Quivers, desingularizations and canonical

  14. [22]

    Rouquier; 2-Kac-Moody algebras, Preprint, 2008, arXiv:0812.5023

    R. Rouquier; 2-Kac-Moody algebras, Preprint, 2008, arXiv:0812.5023

    2008 arXiv

  15. [23]

    Rouquier; Quiver Hecke algebras and 2-Lie algebras,

    R. Rouquier; Quiver Hecke algebras and 2-Lie algebras,

  16. [24]

    Saito; PBW basis of quantized universal enveloping algebras, Publ

    Y. Saito; PBW basis of quantized universal enveloping algebras, Publ. Res. Inst. Math. Sci. 30 (1994), no.2, 209-232

    1994

  17. [25]

    Shoji and Z

    T. Shoji and Z. Zhou; Diagram automorphisms and quantum groups, J. Math. Soc. 72 , (2020), 639-671

    2020

  18. [26]

    Shoji and Z

    T. Shoji and Z. Zhou; Diagram automorphisms and canonical bases for quantum affine algebras, J. Algebra, 569 (2021), 67-110

    2021

  19. [27]

    Varagnolo and E

    M. Varagnolo and E. Vasserot; Canonical bases and KLR-algebras,

  20. [28]

    Xi; A commutation formula for root vectors in quantized enveloping algebras, Pacific J

    N. Xi; A commutation formula for root vectors in quantized enveloping algebras, Pacific J. of Math. 189 (1999), 179-199

    1999

  21. [29]

    Xi; Canonical bases for type B_2 , J

    N. Xi; Canonical bases for type B_2 , J. Algebra, 218 (1999), 8-21

    1999

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.