Pith. sign in

REVIEW 3 major objections 5 minor 18 references

Jones--Wenzl projections of type $D$ and Dyck tilings

T0 review · 3 major / 5 minor · reviewed 2026-08-11 · deepseek-v4-flash

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

Pith's one-line read Every coefficient in the type D Jones–Wenzl projection is a non-recursive weighted count of Dyck tilings.

desk verdict Genuinely new odd-dot type D coefficient formula via bi-colored Hermite histories, but the central theorem leans on an unproved (Q4)/(Q4') equivalence that needs a real proof before I'd fully trust it. read the letter →

arxiv 2412.16966 v1 pith:NUWZBX4H submitted 2024-12-22 math.CO math-phmath.MPmath.QAmath.RT

Keiichi Shigechi This is my paper · ORCID
classification math.COmath-phmath.MPmath.QAmath.RT MSC 05A1505E10
keywords Jones-WenzlprojectionstypeDTemperley-LiebalgebraDycktilingsbi-coloredverticalHermitehistoriesgeneratingfunctionsq-integerscover-inclusivediagrams
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

This paper proves that the coefficients of the Jones–Wenzl projection in the type D Temperley–Lieb algebra are enumerative: each coefficient can be written as a weighted generating function of Dyck tilings. For diagrams with an even number of dots, the coefficient is a Dyck-tiling generating function of the same kind already used for types A and B. The genuinely new result covers diagrams with a single dot, where the product $E_0E_1$ appears; the paper removes the minus signs from the recursion by rewriting it through a type A combination and then interprets every term as a tiling history satisfying conditions (Q1) to (Q4). If correct, this gives a non-recursive, cancellation-free description of all coefficients of the type D projection, making the combinatorial content of each term transparent.

What carries the argument

The central mechanism is the bi-colored vertical Hermite history on a cover-inclusive Dyck tiling: vertical trajectories are colored red or green, with exactly one green trajectory of non-zero length, only red trajectories to its left, and no colored trajectories to its right. Condition (Q4) requires the numbers $n_i$ of tiles in the trajectories to the right of the green trajectory to be weakly decreasing from left to right, which the proof identifies with the post-deletion condition (Q4') that after $m$ deletions the remaining tiling has only trivial Dyck tiles. These histories are exactly what the sign-free recursion (3.12) needs: red trajectories carry the repeated $g_{n,i}$-removals, the unique green trajectory marks the $h_{n,j}$-factor, and the uncolored region contributes the type A generating function.

What would settle it

One concrete check is to enumerate all Dyck tilings of a fixed small size, say $n=5$, with bottom path $\lambda$ and top path $\mu_0$, and test for every choice of green trajectory whether condition (Q4) and condition (Q4') are equivalent; a single tiling satisfying one but not the other would break Theorem 4.16. An independent cross-check is to compute the coefficient of $E_0E_1E_3$ in $Q_4$ from the recurrence (3.2) and from $Z'(\lambda,\mu_0)$; any mismatch would falsify the theorem.

Watch

Extended reading notes

Core claim

The paper's central claim is Theorem 4.16: for a type D diagram $D^{\mathrm{odd}}$ with a single dot and underlying Dyck path $\lambda$, the coefficient of $D^{\mathrm{odd}}$ in the Jones–Wenzl projection $Q_n$ equals $Z'(\lambda,\mu_0)$, the generating function over Dyck tilings above $\lambda$ and below the top path $\mu_0$, where each tiling carries a bi-colored vertical Hermite history satisfying conditions (Q1) to (Q4). Together with Theorem 4.9, which handles even-dot diagrams by the generating function $Z(\mu)$, this puts every coefficient of $Q_n$ in non-recursive enumerative form: the coefficient is a sum of explicit tiling weights rather than a recursively defined linear combination.

Load-bearing premise

The proof depends on the equivalence between condition (Q4), that the tile counts $n_i$ of the deleted trajectories weakly decrease, and condition (Q4'), that after $m$ deletions the remaining tiling contains only trivial Dyck tiles; the paper asserts this equivalence is obvious in Remark 4.13 rather than proving it.

Editorial extensions

If this is right

  • All coefficients of the type D Jones–Wenzl projection, odd-dot and even-dot alike, have a single non-recursive combinatorial expression as weighted sums over Dyck tilings.
  • The odd-dot case, which is the only case where the product $E_0E_1$ is present, is brought onto the same enumerative footing as the even-dot case, so the parity split in the projection becomes a split between two tiling generating functions.
  • Explicit closed forms such as $\operatorname{coef}_n(E_0E_1)=[n]^3/([2n][n+1])$ follow from this framework and can be checked directly from the tiling sum.
  • Because the recursion (3.12) is sign-free, each individual tiling history contributes to a coefficient without cancellations, making the combinatorial meaning of every term transparent.

Reading between the lines

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

  • The monotonicity condition (Q4) resembles the kind of constraint that often yields a bijection to partitions or standard Young tableaux; if such a bijection exists, the coefficient formulas in Section 5 would acquire purely combinatorial proofs independent of the Temperley–Lieb recursion.
  • The bi-colored history construction may extend to other generalized Temperley–Lieb algebras whose diagrams carry a distinguished generator, provided an analogue of the parity reduction to type A can be found.
  • One can test directly whether each coefficient, after clearing denominators, is a polynomial with non-negative coefficients; the cancellation-free tiling sum is a natural setting to look for such positivity.
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

3 major / 5 minor

Summary. The paper proves that every coefficient of the Jones–Wenzl projection in the Temperley–Lieb algebra of type D can be written as a non-recursive generating function over Dyck tilings. The proof strategy is to start from Sentinelli's recurrence for type D, simplify it into two diagrammatic recurrences depending on the parity of the number of dots, remove the minus signs in the odd-dot case by comparing with type A, and then translate the resulting recurrence into a Dyck-tiling enumeration using a new object called bi-colored vertical Hermite histories. The main results are Theorem 4.9 for diagrams with an even number of dots, where the coefficient equals a Dyck-tiling generating function Z(μ), and Theorem 4.16 for diagrams with a unique dot, where the coefficient equals a generating function Z'(λ, μ0) over bi-colored vertical Hermite histories. Section 5 derives explicit closed forms for several families of coefficients, and Appendix A lists the full projection Q3.

Significance. If the central results are correct, the paper gives the first non-recursive, sign-free, enumerative description of all coefficients of type D Jones–Wenzl projections, extending the author's earlier type A and B results and going beyond Baine's Kazhdan–Lusztig interpretation, which explicitly excludes type D. The paper is well organized, includes worked examples that check the formulas against explicit recurrences, and uses no fitted parameters, so the claimed enumerations are genuinely parameter-free. The new bi-colored vertical Hermite histories appear to be a useful combinatorial tool. However, two load-bearing steps are currently not proved in sufficient detail: the algebraic solution of the coefficient recurrences in Proposition 3.3 and, more seriously, the equivalence between conditions (Q4) and (Q4') asserted in Remark 4.13 and used in the proof of Theorem 4.16. Because the main theorem depends on this equivalence, the manuscript needs a substantial revision before the enumerative claim can be accepted.

major comments (3)
  1. [§3.2, proof of Proposition 3.3] The proof of Proposition 3.3 says, after Eq. (3.5), that 'these recursive formulas can be solved as' the coefficient formulas in (3.3), but neither the recursive formulas themselves nor the solution steps are displayed. These coefficient formulas are the algebraic foundation for every later recurrence, including Propositions 3.7, 3.10 and 3.15 and hence Theorems 4.9 and 4.16. Please write out the recursive equations for coef'(g_{n,i}), coef'(h_{n,j}), coef(g_{n,i}) and coef(h_{n,j}), and show the q-integer identities used to solve them, either in the text or in an appendix.
  2. [§4.5, Definition 4.11 and Remark 4.13] The claimed equivalence between condition (Q4) and condition (Q4') is asserted, not proved. Remark 4.13 gives only the observation that n_i < n_{i+1} creates a non-trivial Dyck tile; it does not prove that every non-trivial Dyck tile remaining after the deletion operations forces some increase n_i < n_{i+1}, nor does it establish the converse direction after the deletion and shrinking steps. This equivalence is load-bearing: the proof of Theorem 4.16 uses it to conclude that the set of bi-colored vertical Hermite histories summed in Z'(λ, μ0) is exactly the set produced by the recurrence (3.12). If the equivalence fails in either direction, the equality in Eq. (4.4) does not follow. A complete proof of the equivalence, or a reformulation that avoids it, is required.
  3. [§4.5, proof of Theorem 4.16] The proof of Theorem 4.16 is a high-level sketch in several places. In particular, it is not formally shown that the product-of-weights over a red or green trajectory in (4.3) reproduces the corresponding coefficient factors in (3.12), nor that the type-A region generating function (4.2) is exactly the sum over all tilings of the region to the right of the green trajectory. The discussion of the factor 2^{N(tr)} is also terse: the paper states that a factor 2 arises when i=1∈I(D), but does not explain why exactly the red trajectories with h_b(tr)=1 receive this factor and no others. Please give a detailed bijection between the terms of the recurrence (3.12) and the components of the bi-colored vertical Hermite history, or at least make the weight-to-coefficient correspondence explicit enough to be checked.
minor comments (5)
  1. [§1 and throughout] There is a recurring typo 'Tempereley–Lieb' (for example in the second paragraph of Section 1); it should be 'Temperley–Lieb'.
  2. [Definition 4.8] In the definition of Z(μ), the condition on the excluded tilings is stated as 'there is no Dyck tile of size l(c) above the dotted cap c'; please clarify that this is required for every dotted cap c, and explain what 'above' means precisely when several dotted caps are present.
  3. [Example 4.18] The eight admissible bi-colored vertical Hermite histories are displayed without numbering, and the long weight expression that follows is hard to match to the individual histories. Adding labels to the diagrams and to the summands would improve readability.
  4. [§4.4, paragraph beginning 'The second observation'] The sentence 'Some Dyck tilings are not allowed in type B, and such Dyck tilings are also not allowed in type D as well' is vague; since Theorem 4.9 depends on this restriction, please spell out the exact condition in terms of Dyck tiles.
  5. [Proposition 4.6] In the proof of part (c), the word 'legnth' should be 'length'.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the type-D coefficient formula is derived from the external Sentinelli recurrence and is not reduced to its own input.

full rationale

The derivation chain is self-contained with respect to circularity. The main result Theorem 4.16 is proven by showing that the generating function Z'(λ, μ0) over bi-colored vertical Hermite histories satisfies recurrence (3.12), which is obtained from Sentinelli's external recurrence [12] (Proposition 3.2) via Propositions 3.3, 3.7, 3.10, 3.13 and 3.15. The weights in Definition 4.15 are chosen to match the coefficients coef(hn,i), coef(gn,i), and the type-A weight [h]/([h+1]); the proof verifies the recurrence rather than assuming the target coefficient. The author's earlier work [13] is used only for analogy with types A/B and for the same algorithm, but Theorem 4.9 gives its own proof, and the odd-dot case does not assume the conclusion. The nearest weakness is Remark 4.13: the equivalence between (Q4) and (Q4') is asserted with 'it is obvious' and a brief explanation, and this equivalence is load-bearing for the (Q4) part of Theorem 4.16. That is an unproved combinatorial claim, a correctness/detail risk, but not circular: neither side of the equivalence is the theorem's conclusion, and the theorem would not collapse by definitional identity if the equivalence failed; it would require a different proof. No fitted constants, no prediction of fitted data, and no renaming of a known result are present.

Assumptions & free parameters 0 free parameters · 3 assumptions · 0 invented entities

The central claim is not tuned to data and uses no fitted parameters. It depends on established Temperley-Lieb and Dyck tiling facts plus one paper-specific combinatorial equivalence, the Q4/Q4' assertion, that is the most fragile input.

assumptions (3)
  • domain assumption The type D Jones-Wenzl projection Q_n is unique and satisfies the Sentinelli recurrence Q_n = Q_{n-1} + A_n Q_{n-1} E_n Q_{n-1} + B_n Q_{n-1} E_{w_n} Q_{n-1}.
    This is Proposition 3.2, taken from reference [12], and all later recurrences in the paper derive from it.
  • domain assumption Cover-inclusive Dyck tilings, horizontal and vertical Hermite histories, and their trajectory properties from references [10] and [14] are valid.
    Section 4 builds the generating functions on these existing objects and relies on the stated trajectory properties, notably Proposition 4.6.
  • ad hoc to paper Condition (Q4) is equivalent to condition (Q4'), that repeated deletion of trajectories leaves a tiling consisting only of trivial Dyck tiles.
    This equivalence is asserted after Definition 4.11 and used in the proof of Theorem 4.16. The written justification in Remark 4.13 is brief and informal.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Jones--Wenzl projections of type $D$ and Dyck tilings." pith.science (2026). https://pith.science/paper/NUWZBX4H

@misc{pith2026241216966,
  author       = {Pith},
  title        = {Pith review of: Jones--Wenzl projections of type $D$ and Dyck tilings},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/NUWZBX4H}},
  note         = {Machine review of arXiv:2412.16966}
}
abstract

We study the relation between a coefficient of an element of the Jones--Wenzl projection in the Temperley--Lieb algebra of type $D$ and an enumeration of Dyck tilings. The coefficient can be non-recursively expressed as an enumerative generating function of Dyck tilings by considering the generalized Hermite histories, which we call bi-colored vertical Hermite histories, on the tilings.

Figures

Figures reproduced from arXiv: 2412.16966 by the authors.

Figure 2.1
Figure 2.1. An example of 8-strand Temperley–Lieb diagram three cups. The cup connecting the first and second points is both inner-most and outer-most. The cup connecting the fourth and seventh points is neither inner-most nor outer-most. The cup connecting the fifth and sixth points is inner-most. We have two vertical strands which connect points on the bottom and on the top. The left-most vertical strand is an outer-most cap.… view at source ↗
Figure 3.5
Figure 3.5. The graphical representation of g3,i and h3,j . We recall the recurrence relation via diagrams for TLA n+1. In [11] (see also [4]), Morrison gave a recursive formula for the coefficients of the elements, equivalently for n + 1-strand Temperley–Lieb diagrams. By definition, note that the vertical strand which connects the right-most points on the top and on the bottom is called an inner-most cap. Any n + 1-strand Tem… view at source ↗
Figure 4.1
Figure 4.1. An example of a cover-inclusive Dyck tiling of size 8 tiling in [PITH_FULL_IMAGE:figures/full_fig_p015_4_1.png] view at source ↗
Figures from the paper (3 more)
Figure 4.4
Figure 4.4. Figure 4.4: A Dyck path corresponding to a 4-strand Temperley–Lieb diagram. We call a graph consisting of 2n points and n non-crossing caps with or without dots a chord diagram. The number of chord diagrams consisting of n caps without dots is given by the Catalan number Cn. Sim…
Figure 4.5
Figure 4.5. Figure 4.5: Horizontal Hermite history (left) and vertical Hermite history (right). draw a line in Dyck tiles in a Dyck tiling as follows. The boundary of a Dyck tile consists of up steps ( ) and down steps ( ). For a horizontal Hermite history, we write a line from the left-mos…
Figure 4.12
Figure 4.12. Figure 4.12: A deletion of a Dyck tiling deletion of T. We perform the deletion of T m times. Then, we obtain a Dyck tiling T ′ of size n − m. The condition (Q4) is equivalent to the following condition on T ′ : (Q4′ ) T ′ consists of only trivial Dyck tiles [PITH_FULL_IMAGE:fi…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

18 extracted references · 15 canonical work pages

  1. [2]

    Elias and G

    B. Elias and G. Williamson, Soergel Calculus, Represent. Theory 20 (2016), 295–374, arXiv:1309.0865, doi

    arXiv 2016

  2. [1]

    On the coefficients in the Jones-Wenzl idempotent

    J. Baine, On the coefficients in the Jones–Wenzl idemotent , preprint (2024), 13 pages, arXiv:2406.06333

    work page Pith review arXiv 2024

  3. [3]

    C. K. Fan, A Hecke algebra quotient and some combinatorial applicatio ns, J. Algebraic Combin. 5 (1996), no. 3, 175–189, doi

    work page 1996

  4. [4]

    I. B. Frenkel and M. G. Khovanov, Canonical bases in tensor products and graphical calculus f or Uq(sl2), Duke Math. J. 87 (1997), no. 3, 409–480

    work page 1997

  5. [5]

    R. M. Green, Generalized Temperley–Lieb algebras and decorated tangles, J. Knot Theory Ramifications 7 (1998), 155–171, arXiv:q-alg/9712018

    work page Pith review arXiv 1998

  6. [6]

    V. F. R. Jones, Index for subfactors , Invent. Math. 72 (1983), 1–25

    work page 1983

  7. [7]

    , Hecke algebra representations of braid groups and link poly nomials, Ann. of Math. 126 (1987), no. 2, 335–388

    work page 1987

  8. [8]

    , On knot invariants related to some statistical mechanical m odels, Pacific J. Math. 137 (1989), no. 2, 311–334

    work page 1989

Show all 18 references
  1. [9]

    R. W. Kenyon and D. B. Wilson, Double-dimer pairings and skew Young diagrams , Electron. J. Combin. 18 (2011), no. 1, P130, arXiv:1007.2006

    2011 arXiv

  2. [10]

    J. S. Kim, K. M´ esz´ aros, G. Panova, and D. B. Wilson, Dyck tilings, increasing trees, descents and inversions , J. Combin. Theory Ser. A 122 (2014), 9–27, arXiv:1205.6578

    2014 arXiv

  3. [11]

    Morrison, A formula for the Jones–Wenzl projections , Proc

    S. Morrison, A formula for the Jones–Wenzl projections , Proc. Centre Math. Appl. (2017), 367–378, arXiv:1503.00384

    2017 arXiv

  4. [12]

    Sentinelli, The Jones-Wenzl idempotent of a generalized Temperley-Lie b algebra , Journal of Algebra 528 (2019), 505–524, doi

    P. Sentinelli, The Jones-Wenzl idempotent of a generalized Temperley-Lie b algebra , Journal of Algebra 528 (2019), 505–524, doi

    2019

  5. [13]

    Shigechi, Jones–Wenzl projections and Dyck tilings: type A and B, preprint (2024), 12 pages, arXiv:2402.09887

    K. Shigechi, Jones–Wenzl projections and Dyck tilings: type A and B, preprint (2024), 12 pages, arXiv:2402.09887

    2024 arXiv

  6. [14]

    Shigechi and P

    K. Shigechi and P. Zinn-Justin, Path representation of maximal parabolic Kazhdan–Lusztig polynomials, J. Pure Appl. Alegebra 216 (2012), no. 11, 2533–2548, arXiv:1001.1080

    2012 arXiv

  7. [15]

    J. R. Stembridge, On the Fully Commutative Elements of Coxeter Groups , J. Algebr. Comb. 5 (1996), 353–385, doi

    1996

  8. [16]

    Stroppel and Z

    C. Stroppel and Z. Wojciechowski, Diagrammatics for the smallest quantum coideal and Jones–W enzl projectors, preprint (2024), 20 pages, arXiv:2406.12132

    2024 arXiv

  9. [17]

    H. N. V. Temperley and E. H. Lieb, Relations between the ‘percolation ’ and ‘coloring’ problem and other graph- theoretical problems associated with regular planar latti ces: some exact results for the ‘percolation problem’ , Proc. Roy. Soc. Lond. A 322 (1971), 251–280

    1971

  10. [18]

    Wenzl, On sequences of projections , C

    H. Wenzl, On sequences of projections , C. R. Math. Rep. Acad. Sci. Canada 9 (1987), no. 1, 5–9. Email address : k1.shigechi AT gmail.com

    1987

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.