Bounds on skew dimensions and characters of symmetric groups via thick hook decompositions
Lucas Teyssier · 2025
Read Bounds on skew dimensions and characters of symmetric groups via thick hook decompositions by Lucas Teyssier (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2508-06770.txt.
Introduction
Lucas Teyssier finds sharper bounds on hard counting problems for symmetric groups by cutting Young diagrams into thick hooks. These diagrams encode representations, and a known formula counts their standard fillings through hook lengths. His 2025 paper controls those products better for skew shapes, improving bounds on how many fillings a shape can have.
Joined with basic counting inside a classic character rule, the bounds give a uniform cap on character values across the whole symmetric group. For balanced representations the cap beats earlier results when permutations move many points, and it is sharp at the largest scale. Short proofs also recover known bounds on Kronecker coefficients, quantities that resist direct attack. The work links diagram geometry to representation size.
Catalog record
| CREATOR | Lucas Teyssier |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 53 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2508-06770
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Bounds on skew dimensions and characters of symmetric groups via thick hook decompositions Lucas Teyssier Abstract We introduce a new decomposition of Young diagrams into thick hooks. These thick hook decompositions enable having a better control over hook products for excited diagrams in the Naruse hook length formula, which leads to improved bounds on the number of standard tableaux of skew shapes. Combining this with elementary counting arguments in the Murnaghan–Nakayama rule, we establish a uniform bound on characters of symmetric groups 𝔖 n \mathfrak{S}_{n} . In the case of balanced representations, this improves on the character bounds of Féray and Śniady for permutations with support size at least n 2 / 3 n^{2/3} , and is sharp for permutations with support size of order n n . Finally, we recover some bounds of Pak and Panova on Kronecker coefficients with very short proofs that rely on characters. 1 Introduction Let \lambda be an irreducible representation of {\mathfrak{S}}_{n} and \sigma\in{\mathfrak{S}}_{n} . Denote the dimension of \lambda by d_{\lambda} , the character of \lambda at \sigma by \ch^{\lambda}(\sigma) , the associated renormalized character by \chi^{\lambda}(\sigma):=\ch^{\lambda}(\sigma)/d_{\lambda} , the minimal number of transpositions one needs to multiply to obtain \sigma by |\sigma| , and the support size of \sigma by \supp(\sigma) . The irreducible representations of {\mathfrak{S}}_{n} are in bijection with integer partitions \lambda\vdash n , which can be represented by Young diagrams, and we use the same notation \lambda for these three notions. The first row (resp. column) of a diagram \lambda is denoted by \lambda_{1} (resp. \lambda^{\prime}_{1} ), and the maximal hook length of \lambda is s(\lambda):=\lambda_{1}+\lambda_{1}^{\prime}-1 . Finally, for C>0 we say that \lambda\vdash n is C -balanced if s(\lambda)\leq C\sqrt{n} . Building on the work of Kerov [ 8 ] , Biane [ 1 , 2 ] proved, alongside factorization formulas for characters, that for balanced diagrams \lambda and permutations \sigma with \supp(\sigma)=O(1) , |\chi^{\lambda}(\sigma)|\leq\left(\frac{O(1)}{\sqrt{n}}\right)^{|\sigma|}. (1.1) This bound on characters for balanced diagrams was later extended to \supp(\sigma)=O(\sqrt{n}) by Rattan and Śniady [ 27 ] . Moore and Russell conjectured ( [ 27 , Conjecture 3] ) that ( 1.1 ) holds for any support size. Note that we always have \supp(\sigma)/2\leq|\sigma|\leq\supp(\sigma)-1 , so \supp(\sigma)\asymp|\sigma| . The known bounds for \supp(\sigma)=\omega(\sqrt{n}) are however still relatively weak, especially for larger support sizes. Rattan and Śniady [ 27 ] proved that for permutations with \supp(\sigma)=\omega(\sqrt{n}) , |\chi^{\lambda}(\sigma)|\leq\left(O(1)\frac{|\sigma|^{2}}{n^{3/2}}\right)^{|\sigma|}, (RŚ) which is trivial if \supp(\sigma)=\omega(n^{3/4}) . The strongest known character bound in this setting is due to Féray and Śniady [ 5 ] , who proved the following bound, which applies to all shapes of diagrams: |\chi^{\lambda}(\sigma)|\leq\left(O(1)\max\left(\frac{|\sigma|}{n},\frac{s(\lambda)}{n}\right)\right)^{|\sigma|}. (FŚ) For balanced diagrams, ( FŚ ) coincides with the bound of Rattan and Śniady if |\sigma|=O(\sqrt{n}) ; these bounds are excellent and led to optimal applications to the number of standard diagrams of balanced skew Young tableaux in [ 4 ] . If |\sigma|=\omega(\sqrt{n}) , and still for balanced diagrams \lambda , ( FŚ ) can be rewritten as |\chi^{\lambda}(\sigma)|\leq\left(O(1)\frac{|\sigma|}{n}\right)^{|\sigma|}. (FŚbal) This bound is stronger than ( RŚ ) since |\sigma|/n=o(|\sigma|^{2}/n^{3/2}) . Our main result is the following character bound. Theorem 1.1 . As n\to\infty , uniformly over all representations \lambda\vdash n and all permutations \sigma\in{\mathfrak{S}}_{n}\backslash\left\{\Id\right\} , we have |\chi^{\lambda}(\sigma)|\leq\left(\frac{O(1)}{\sqrt{|\sigma|}}\right)^{|\sigma|}\max\left(1,\frac{s(\lambda)\sqrt{|\sigma|}}{n}\right)^{\supp(\sigma)}. (1.2) Theorem 1.1 has the following simpler form for balanced representations. Theorem 1.2 . Let C>0 . As n\to\infty , uniformly over all C -balanced representations \lambda\vdash n and all permutations \sigma\in{\mathfrak{S}}_{n}\backslash\left\{\Id\right\} , we have |\chi^{\lambda}(\sigma)|\leq\left(\frac{O(1)}{\sqrt{|\sigma|}}\right)^{|\sigma|}. (1.3) An interesting aspect of our results is that they are more precise for permutations with larger support sizes, contrary to the bounds of [ 27 ] and [ 5 ] . More precisely, Theorem 1.2 is stronger than ( FŚbal ) if |\sigma|=\omega(n^{2/3}) , but weaker if |\sigma|=o(n^{2/3}) . As such, Theorem 1.1 complements the Féray–Śniady character bounds, as we illustrate in Figure 1 . Moreover, Theorem 1.2 agrees with the conjecture of Moore and Russell for permutations with \supp(\sigma)\asymp n . Theorems 1.1 and 1.2 are proved in Section 3.3 . To this end, we establish two intermediate results, which are of independent interest. The first one is a bound on the number d_{\lambda\backslash\mu} of standard tableaux of a skew diagram \lambda\backslash\mu , in function of the maximal hook length s(\lambda) of \lambda and the sizes of \lambda and \mu . Theorem 1.3 . There exists a universal constant C>0 such that the following holds. Let n\geq 1 and 1\leq k\leq n be integers. Let \lambda\vdash n and \mu\vdash k such that \mu\subset\lambda . Then \frac{d_{\lambda\backslash\mu}}{d_{\lambda}}\leq\left(C\max\left(\frac{1}{\sqrt{k}},\frac{s(\lambda)}{n}\right)\right)^{k}. (1.4) Theorem 1.3 is proved in Section 2 . It relies on thick hook decompositions in the Naruse hook length formula. We also show in Section 2.9 that Theorem 1.3 is sharp for all orders of magnitude for k and s(\lambda) . Our second intermediate result is a simple character bound, written in terms of the number of rows \ell(\lambda) – or the diagonal length \delta(\lambda) – of a diagram \lambda and the number of cycles \cyc(\sigma) of a permutation \sigma . Theorem 1.4 . Let n\geq 1 , \lambda\vdash n , and \sigma\in{\mathfrak{S}}_{n} . Then (a) |\ch^{\lambda}(\sigma)|\leq\ell(\lambda)^{\cyc(\sigma)} ; (b) |\ch^{\lambda}(\sigma)|\leq(2\delta(\lambda))^{\cyc(\sigma)} . Theorem 1.4 is proved in Section 3.2 . Its proof relies on the Murnaghan–Nakayama rule and elementary counting arguments. We also use it in Section 5 to obtain bounds on Kronecker coefficients. \displaystyle y=\frac{\ln(B^{-1/|\sigma|})}{\ln n} \displaystyle x=\frac{\ln|\sigma|}{\ln n} 1/2 2/3 {1} 1/3 {1/2} 0 Conjectured bound (Moore–Russell) Previous best bound (Féray–Śniady) Our bound from Theorem 1.2 Figure 1: Asymptotic uniform character bounds for balanced diagrams, logarithmically rescaled. Here B denotes the upper bound on \left|\chi^{\lambda}(\sigma)\right| from the right-hand-side of ( 1.1 ) for the conjecture, ( FŚbal ) for Féray–Śniady, and ( 1.3 ) for our bound. The implicit universal constants (in the definition of balanced diagrams and the character bounds) have a negligible asymptotic contribution, with the rescaling of the figure. Notation We use the following asymptotic notation. We write f(n)=O(g(n)) or f(n)\lesssim g(n) or g(n)=\Omega(f(n)) if there exists a constant C>0 such that |f(n)|\leq C|g(n)| for all n large enough, f(n)=o(g(n)) or g(n)=\omega(f(n)) if f(n)/g(n)\xrightarrow[n\to\infty]{}0 , and f(n)\asymp g(n) if f(n)\lesssim g(n) and f(n)\gtrsim g(n) . We use the French convention for Young diagrams. 2 Bounds on skew dimensions Let \lambda and \mu\subset\lambda be Young diagrams. We denote the number of standard tableaux of the skew diagram \lambda\backslash\mu by d_{\lambda\backslash\mu} , and refer to this quantity as skew dimension . Different explicit formulas have been found to compute skew dimensions, see for instance [ 23 , 29 ] . A decade ago, Naruse [ 22 ] found a way to compute skew dimensions via excited diagrams, a notion introduced in [ 6 ] . This powerful formula, now called Naruse hook length formula, has seen other proofs and generalizations, see [ 18 , 11 ] , and has enabled computing improved bounds on skew dimensions in [ 17 , 20 , 24 ] as well as explicit formulas for various skew shapes [ 9 , 19 , 10 ] . In this section we are concerned in proving general bounds on the quantity \frac{d_{\lambda\backslash\mu}}{d_{\lambda}} for all shapes, in function of the sizes of the diagrams \lambda and \mu and the maximal hook length of \lambda . 2.1