betaFeedback

WorksMathematics

The big cactus group, invariant tensors, and rectangular tableaux

Abstract

The cactus group acts on tensor products of crystals and, through Bender–Knuth involutions, on semistandard Young tableaux. We show that the big cactus group Jn′=π1Sn(M‾0,n(R))J'_n=π_1^{\mathfrak S_n}(\overline M_{0,n}(\mathbb R)), n≥3n\ge3, is obtained from the cactus group by imposing n−1n-1 Dynkin-reversal relations, which say that conjugation by the full reversal acts on the Bender–Knuth generators by the Dynkin diagram automorphism; the quotient map is induced by forgetting the distinguished marked point. We give a second presentation, in which these relations are replaced by the requirement that every reduced word for the longest element of Sn\mathfrak S_n adapted to an orientation of An−1A_{n-1} represents the full reversal. These relations also hold on rectangular tableaux, where every adapted reduced word therefore acts as evacuation, and on invariant tensors: for finite normal crystals of any complex semisimple Lie algebra, the cactus action on invariants factors through Jn′J'_n. We prove the latter combinatorially in type AA and in general via Gaudin eigenline coverings. Haiman's promotion relation follows from a single adapted-word relation, but for n≥4n\ge4 it is strictly weaker than the whole family.

Introduction

Bender–Knuth involutions and their group

Bender–Knuth involutions t1,…,tn−1t_1,\ldots,t_{n-1} act on semistandard Young tableaux with entries in {1,…,n}\{1,\ldots,n\}. The involution tit_i rearranges the entries ii and i+1i+1 so as to interchange the number of ii’s and i+1i+1’s, while preserving the shape and all other entries. They were introduced to give a combinatorial proof that Schur functions are symmetric.

The individual involutions tit_i are simple, but the group they generate is not. Berenstein and Kirillov [6] initiated the study of the group BKn\mathrm{BKn} generated by t1,…,tn−1t_1,\ldots,t_{n-1} acting on semistandard tableaux of all shapes, now called the Berenstein–Kirillov group. Two distinguished elements are

promotion pn−1=tn−1tn−2⋯t1p_{n-1}=t_{n-1}t_{n-2}\cdots t_1, evacuation qn−1=t1(t2t1)(t3t2t1)⋯(tn−1⋯t1)q_{n-1}=t_1(t_2t_1)(t_3t_2t_1)\cdots(t_{n-1}\cdots t_1).

Finding a presentation of BKn\mathrm{BK}_n is still open, and appears to be hard.

Chmutov, Glick and the second author [12] showed that the Bender–Knuth involutions satisfy the relations of the cactus group CnC_n of Henriques and Kamnitzer [27], giving a surjection Cn↠BKnC_n \twoheadrightarrow\mathrm{BK}_n. The same conclusion was obtained independently by Halacheva [24] using skew Howe duality, and concurrently by Berenstein and Kirillov [7] by a group-theoretic argument. Thus the cactus group gives a large family of relations in BKn\mathrm{BK}_n that come from a general structure rather than from computation.

Evacuation words on a rectangle

This paper concerns what happens when one restricts the Bender–Knuth action to rectangular shapes, which in type AA model Kirillov–Reshetikhin crystals. Let BKRn\mathrm{BKR}_n be the quotient of BKn\mathrm{BK}_n obtained by imposing all relations that hold on semistandard tableaux of every rectangular shape with entries in {1,…,n}\{1,\ldots,n\}.

Call a reduced word i=(i1,…,iN)\mathbf{i}=(i_1,\ldots,i_N) for w0∈Snw_0\in\mathfrak{S}_n an evacuation word if the corresponding product ti1⋯tiNt_{i_1}\cdots t_{i_N} acts as evacuation. The standard product

qn−1=t1(t2t1)(t3t2t1)⋯(tn−1⋯t1)q_{n-1}=t_1(t_2t_1)(t_3t_2t_1)\cdots(t_{n-1}\cdots t_1)

is an evacuation word, as is every commutation-equivalent reduced word. On general shapes we know of no others, and conjecture that there are none (Conjecture 1.3, verified for n≤5n\leq5). On rectangles, many more words work.

Recall that a reduced word for w0w_0 is adapted to an orientation Ω\Omega of the Dynkin diagram An−1A_{n-1} if i1i_1 is a source of Ω\Omega, i2i_2 is a source of si1Ωs_{i_1}\Omega, and so on. Each reduced word is adapted to at most one orientation, and the words adapted to a fixed Ω\Omega form a single commutation class [34, 5].

Theorem 1.1 (Theorem 5.5). On rectangular tableaux, every reduced word for w0w_0 adapted to an orientation of An−1A_{n-1} is an evacuation word.

Thus the single commutation class coming from the linear orientation is replaced on rectangles by all 2n−22^{n-2} adapted classes. We conjecture that these are all the reduced words representing evacuation on rectangles.

Definition 1.2. A reduced word ww for w0∈Snw_0\in\mathfrak{S}_n is an evacuation word in BKn\mathrm{BK}_n (resp. BKRn\mathrm{BKR}_n) if tw=qn−1t_w=q_{n-1} in BKn\mathrm{BK}_n (resp. BKRn\mathrm{BKR}_n).

Conjecture 1.3. The evacuation words in BKn\mathrm{BK}_n are exactly the reduced words for w0w_0 adapted to the linear orientation.

Here the linear orientation is

1⟶2⟶⋯⟶n−11\longrightarrow2\longrightarrow\cdots\longrightarrow n-1

For the opposite linear orientation, the adapted words represent α(qn−1)\alpha(q_{n-1}), the dual evacuation, on arbitrary tableaux. On rectangles, dual evacuation agrees with evacuation; equivalently, α(qn−1)=qn−1\alpha(q_{n-1})=q_{n-1}. Thus Theorem 1.1 extends this identity from the opposite linear orientation to every orientation of An−1A_{n-1}.

Conjecture 1.4. The evacuation words in BKRn\mathrm{BKR}_n are exactly the reduced words for w0w_0 adapted to some orientation of An−1A_{n-1}.

We verified both conjectures by computer for n≤5n\leq5.

Equivalent relations

The exponentially large family of evacuation identities from Theorem 1.1 is equivalent to a very small set of relations. Let α\alpha be the index-reversing map α(ti)=tn−i\alpha(t_i)=t_{n-i}. The Dynkin-reversal relations are

qn−1tiqn−1=tn−i,i=1,…,n−1.(1)q_{n-1}t_iq_{n-1}=t_{n-i},\qquad i=1,\ldots,n-1. \tag*{(1)}

In other words, conjugation by evacuation acts on the Bender–Knuth generators by the Dynkin diagram automorphism.

Theorem 1.5 (Theorems 3.4 and 5.4). The relations (∗)(*) hold in BKRn\mathrm{BKR}_n. Moreover, modulo the cactus relations, they are equivalent to the whole family of Theorem 1.1.

Promotion

We also obtain the identity

qn−1α(qn−1)=(t1t2⋯tn−1)n=pn−n+1,q_{n-1}\alpha(q_{n-1})=(t_1t_2\cdots t_{n-1})^n=p_n^{-n+1},

which already holds in the commutation group (Theorem 6.1). For the opposite linear orientation

1←2←⋯←n−1,1\leftarrow2\leftarrow\cdots\leftarrow n-1,

the corresponding adapted relation on rectangles is

α(qn−1)=qn−1.\alpha(q_{n-1})=q_{n-1}.

Together with qn−12=1q_{n-1}^2=1, this gives pn−1n=1p_{n-1}^n=1 on rectangles, which is Haiman’s theorem [23 Thm. 4.4]. We show that for every n≥4n\ge4, the relation pn−1n=1p_{n-1}^n=1 is strictly weaker than the relations (∗)(*) (Proposition 6.4). Thus Haiman’s relation captures only part of the relations we describe on rectangles.

Adapted words and sorting networks

A reduced word for w0w_0 can be viewed as a sorting network on nn wires. Choosing four wires and keeping only the crossings between them gives a reduced word for w0∈S4w_0\in S_4, which we call a 4-subnetwork. Warrington [42] introduced this point of view in connection with Sylvester’s four-point problem. Under the Goodman–Pollack correspondence with planar point configurations, the sixteen reduced words for w0∈S4w_0\in S_4 split into twelve corresponding to four points in convex position and four,

X={123212, 321232, 212321, 232123},X=\{123212,\ 321232,\ 212321,\ 232123\},

corresponding to configurations in which one point lies inside the triangle formed by the other three.

Meng [35] studied reduced words avoiding these four subnetworks and asked how many XX-avoiding reduced words for w0∈Snw_0\in S_n there are. The counts reported by Warrington begin

2,12,328,54520,68641152.2,12,328,54520,68641152.

These are exactly the numbers of adapted reduced words that we obtain. In fact, we prove the following.

Theorem 1.6 (Theorem 4.9). Every adapted reduced word for w0w_0 is XX-avoiding.

The rectangular cactus group

Let RCnRC_n, the rectangular cactus group, be the quotient of CnC_n by the Dynkin-reversal relations. For n≥3n\ge3, Theorem 7.2 identifies this quotient with the big cactus group JnJ_n via the map that forgets the root marked point. By construction,

Cn↠RCn↠BKRn.C_n\twoheadrightarrow RC_n\twoheadrightarrow\mathrm{BKR}_n.

The second map need not be an isomorphism: already for n=4n=4, RC4↠BKR4RC_4\twoheadrightarrow\mathrm{BKR}_4 has nontrivial kernel.

Theorem 1.7 (Theorem 3.10). The natural map RC4↠BKR4RC_4\twoheadrightarrow\mathrm{BKR}_4 is not an isomorphism.

Conjecture 1.8 (Conjecture 3.11).

BKRn≅BKn/⟨qn−1tiqn−1tn−i⟩.\mathrm{BKR}_n \cong\mathrm{BK}_n/\langle q_{n-1}t_iq_{n-1}t_{n-i}\rangle.

In other words, we conjecture that the Dynkin-reversal relations generate all additional relations that appear when one restricts from arbitrary shapes to rectangles. This avoids the much harder problem of finding a presentation of BKn\mathrm{BK}_n itself.

Crystals and invariant tensors

The cactus group of Henriques and Kamnitzer [27] acts on tensor products B1⊗⋯⊗BnB_1 \otimes\cdots\otimes B_n of crystals: the generator q[i,j]q_{[i,j]} reverses the factors i,…,ji,\ldots,j using the crystal commutor, built from the Schützenberger involution. We show that the group RCn≅Jn′\mathcal{R}C_n \cong J'_n has a natural meaning in this setting.

An invariant of a crystal is an element on which every eie_i and every fif_i is undefined, equivalently a one-element connected component. We write Inv⁡(B)\operatorname{Inv}(B) for the set of invariants. Since the cactus action is by crystal isomorphisms, it carries invariants to invariants of the tensor product with the factors permuted.

Theorem 1.9. Let B1,…,BnB_1,\ldots,B_n be finite normal crystals for slm\mathfrak{sl}_m. On Inv⁡(B1⊗⋯⊗Bn)\operatorname{Inv}(B_1 \otimes\cdots\otimes B_n), every reduced word for w0w_0 adapted to an orientation of An−1A_{n-1}, read in the elements ti∈Cnt_i \in C_n of Theorem 2.5, acts by

b1⊗⋯⊗bn⟼ξ(bn)⊗⋯⊗ξ(b1),b_1 \otimes\cdots\otimes b_n \longmapsto\xi(b_n) \otimes\cdots\otimes\xi(b_1),

where ξ\xi is the Schützenberger involution. Equivalently, the cactus action on invariants factors through RCnRC_n.

We give a combinatorial proof in type A and, in § 9.5, a topological proof valid for finite normal crystals of any complex semisimple Lie algebra; see Corollary 8.19.

Theorem 5.5 is the special case in which all factors are single-row crystals; the standard-tableau case is obtained when all factors are B(ϖ1)B(\varpi_1). Thus the tableau relations studied above are the simplest instances of a more general phenomenon for invariant tensors.

The big cactus group and the geometry of invariant tensors

The rectangular cactus group has a topological interpretation. The ordinary cactus group is the equivariant fundamental group [13, 27]

Cn=π1Sn(M‾0,n+1(R)),C_n=\pi_1^{\mathfrak{S}_n}\left(\overline{M}_{0,n+1}(\mathbb{R})\right),

where Sn\mathfrak{S}_n permutes nn marked points and fixes the remaining point, denoted ∞\infty. Henriques’s big cactus group [26] is

References

  1. [1]O. Angel, A. E. Holroyd, Random subnetworks of random sorting networks, Electron. J. Combin. 17 (2010), #N23. (Proves Warrington’s conjecture [42].)arxiv.org/abs/0911.2519
  2. [2]I. Assem, D. Simson, A. Skowroński, Elements of the Representation Theory of Associative Algebras. Vol. 1, London Math. Soc. Student Texts 65, Cambridge Univ. Press, 2006.
  3. [3]O. Azenhas, M. Tarighat Feller, J. Torres, Symplectic cacti, virtualization and Bernstein–Kirillov groups, J. Comb. Algebra 10 (2026), 53–130.arxiv.org/abs/2207.08446
  4. [4]J. Bandlow, A. Schilling, N. M. Thiéry, On the uniqueness of promotion operators on tensor products of type A crystals, J. Algebraic Combin. 31 (2010), no. 2, 217–251.
  5. [5]R. Bédard, On commutation classes of reduced words in Weyl groups, European J. Combin. 20 (1999), no. 6, 483–505.DOI
  6. [6]A. D. Berenstein, A. N. Kirillov, Groups generated by involutions, Gelfand–Tsetlin patterns, and combinatorics of Young tableaux, Algebra i Analiz 7 (1995), no. 1, 92–152; St. Petersburg Math. J. 7 (1996), no. 1, 77–127.
  7. [7]A. D. Berenstein, A. N. Kirillov, Cactus group and Gelfand–Tsetlin group, RIMS preprint RIMS-1858, 2016.
  8. [8]A. D. Berenstein, A. V. Zelevinsky, Canonical bases for the quantum group of type A_r and piecewise-linear combinatorics, Duke Math. J. 82 (1996), 473–502.DOI
  9. [9]M. Borodin, The orbits of the action of the cactus group on arc diagrams, arXiv:2312.01176.
  10. [10]D. Brown, B. Elek, I. Halacheva, Cacti, toggles, and reverse plane partitions, arXiv:2412.02614.arxiv.org/abs/2412.02614
  11. [11]J. H.-H. Chiang, A. T. N. Hoang, M. Kendall, R. Lynch, S. Nguyen, B. Przybocki, J. Xia, Bender–Knuth involutions on linear extensions of posets, Discrete Math. 347 (2024), no. 9, Paper No. 114068.arxiv.org/abs/2302.12425
  12. [12]M. Chmutov, M. Glick, P. Plyavskyy, The Berenstein–Kirillov group and cactus groups, J. Comb. Algebra 4 (2020), no. 2, 111–140.arxiv.org/abs/1609.02046
  13. [13]M. Davis, T. Januszkiewicz, R. Scott, Fundamental groups of blow-ups, Adv. Math. 177 (2003), no. 1, 115–179; numbering follows arXiv:math/0203127v2.DOI
  14. [14]A. Dranowski, B. Elek, J. Kamnitzer, C. Morton-Ferguson, Heaps, crystals, and preprojective algebra modules, Selecta Math. (N.S.) 30 (2024), no. 5, Paper No. 94.
  15. [15]D. Einstein and J. Propp, Piecewise-linear and birational toggling, Discrete Math. Theor. Comput. Sci. Proc. AT (2014), 513–524.arxiv.org/abs/1404.3455
  16. [16]B. Fontaine, J. Kamnitzer, Cyclic sieving, rotation, and geometric representation theory, Selecta Math. (N.S.) 20 (2014), no. 2, 609–625.arxiv.org/abs/1212.1314
  17. [17]G. Frieden, Affine type A geometric crystal on the Grassmannian, J. Combin. Theory Ser. A 167 (2019), 499–563.arxiv.org/abs/1706.02844
  18. [18]W. Fulton, Young Tableaux, London Math. Soc. Student Texts 35, Cambridge Univ. Press, 1997.DOI
  19. [19]A. Garver, R. Patrias, H. Thomas, Minuscule reverse plane partitions via quiver representations, Selecta Math. (N.S.) 29 (2023), no. 3, Paper No. 37, 48 pp.; numbering of results cited here follows arXiv:1812.08345v3.
  20. [20]J. E. Goodman and R. Pollack, On the combinatorial classification of nondegenerate configurations in the plane, J. Combin. Theory Ser. A 29 (1980), 220–235.DOI
  21. [21]D. Grinberg and T. Roby, Iterative properties of birational rowmotion II: rectangles and triangles, Electron. J. Combin. 22 (2015), no. 3, Paper P3.40.DOI
  22. [22]A. Gurenkova, I. Halacheva, L. Rybnikov, Mirabolic cactus group, skew Young tableaux and monodromy of Bethe vectors, in preparation.
  23. [23]M. Haiman, Dual equivalence with applications, including a conjecture of Proctor, Discrete Math. 99 (1992), no. 1–3, 79–113.
  24. [24]I. Halacheva, Skew Howe duality for crystals and the cactus group, arXiv:2001.02262.
  25. [25]I. Halacheva, J. Kamnitzer, L. Rybnikov, A. Weekes, Crystals and monodromy of Bethe vectors, Duke Math. J. 169 (2020), no. 12, 2337–2419.DOI
  26. [26]A. Henriques, An action of the cactus group, arXiv:0705.3000, 2007.arxiv.org/abs/0705.3000
  27. [27]A. Henriques, J. Kamnitzer, Crystals and coboundary categories, Duke Math. J. 132 (2006), no. 2, 191–216.
  28. [28]A. Ilin, J. Kamnitzer, Y. Li, P. Przytycki, L. Rybnikov, The moduli space of cactus flower curves and the virtual cactus group, arXiv:2308.06880, 2023; numbering follows arXiv:2308.06880v2 (2024).arxiv.org/abs/2308.06880
  29. [29]A. Ilin, J. Kamnitzer, L. Rybnikov, Gaudin models and moduli space of flower curves, Adv. Math. 482 (2025), Paper No. 110616; arXiv:2407.06424, numbering of results cited here follows v2.arxiv.org/abs/2407.06424
  30. [30]J. Kamnitzer, L. Rybnikov, Cactus flower spaces and monodromy of Bethe vectors, arXiv:2507.12829, 2025.arxiv.org/abs/2507.12829
  31. [31]V. Krylov, I. Mashanova-Golikova, L. Rybnikov, Bethe subalgebras in Yangians and Kirillov–Reshetikhin crystals, arXiv:2212.11995, 2022; numbering follows arXiv:2212.11995v2 (2026).arxiv.org/abs/2212.11995
  32. [32]A. Lascoux and M.-P. Schützenberger, Le monoïde plaxique, in Noncommutative Structures in Algebra and Geometric Combinatorics, Quaderni de “La Ricerca Scientifica”, vol. 109, Consiglio Nazionale delle Ricerche, Rome, 1981, 129–156.
  33. [33]S. Liao, L. Rybnikov, Maximal transitivity of the cactus group on standard Young tableaux, arXiv:2506.16561.arxiv.org/abs/2506.16561
  34. [34]G. Lusztig, Canonical bases arising from quantized enveloping algebras, J. Amer. Math. Soc. 3 (1990), no. 2, 447–498.
  35. [35]D. Meng, Reduced decompositions and commutation classes, arXiv:1009.0886.arxiv.org/abs/1009.0886
  36. [36]R. Patrias, O. Pechenik, Tableau evacuation and webs, Proc. Amer. Math. Soc. Ser. B 10 (2023), 341–349.
  37. [37]S. Pfannerer, M. Rubey, B. W. Westbury, Promotion on oscillating and alternating tableaux and rotation of matchings and permutations, Algebr. Comb. 3 (2020), no. 1, 107–141.arxiv.org/abs/1804.06736
  38. [38]B. Rhoades, Cyclic sieving, promotion, and representation theory, J. Combin. Theory Ser. A 117 (2010), 1385–1407.
  39. [39]I. Rodrigues, A shifted Berenstein–Kirillov group and the cactus group, arXiv:2104.11799.arxiv.org/abs/2104.11799
  40. [40]R. P. Stanley, Enumerative Combinatorics, Vol. 2, Cambridge Univ. Press, 1999.
  41. [41]R. P. Stanley, Promotion and evacuation, Electron. J. Combin. 16 (2009), no. 2, Research Paper R9, 24 pp.
  42. [42]G. S. Warrington, A combinatorial version of Sylvester’s four-point problem, Adv. in Appl. Math. 45 (2010), 390–394.DOI
  43. [43]B. W. Westbury, Invariant tensors and the cyclic sieving phenomenon, Electron. J. Combin. 23 (2016), no. 4, Paper P4.25.arxiv.org/abs/0912.1512
  44. [44]B. W. Westbury, Coboundary categories and local rules, Electron. J. Combin. 25 (2018), no. 4, Paper No. 4.9.arxiv.org/abs/1705.07141
  45. [45]B. W. Westbury, Interpolating between promotion and the long cycle, arXiv:1906.07146.arxiv.org/abs/1906.07146

Paper details

vlnc:W_rtnwmyxy
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 1

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Mathematics

Not recorded for this paper

View statement
Looking for related papers…
Contents