Dimension identities, almost self-conjugate partitions, and BGG complexes for Hermitian symmetric pairs

TL;DR

This paper generalizes classical identities involving almost self-conjugate partitions by introducing a constant arm-leg difference, establishing new dimension identities between modules, and exploring their implications in representation theory and BGG complexes.

Contribution

It introduces a broad family of dimension identities for modules associated with partitions having a fixed arm-leg difference, extending classical symmetric function identities and BGG complex interpretations.

Findings

Established infinite families of dimension identities between rak{gl}_n and rak{gl}_{n+m} modules.

Identified six infinite families of congruent blocks in parabolic category al, with posets of highest weights related to the partitions.

Generalized Littlewood identities and related BGG complexes based on these new combinatorial structures.

Abstract

An almost self-conjugate (ASC) partition has a Young diagram in which each arm along the diagonal is exactly one box longer than its corresponding leg. Classically, the ASC partitions and their conjugates appear in two of Littlewood's symmetric function identities. These identities can be viewed as Euler characteristics of BGG complexes of the trivial representation, for classical Hermitian symmetric pairs. In this paper, we consider partitions in which the arm-leg difference is an arbitrary constant $m$. By viewing these partitions as highest weights, we establish an infinite family of dimension identities between $\mathfrak{gl}_n$- and $\mathfrak{gl}_{n+m}$-modules. We then interpret this result in the context of blocks in parabolic category $\mathcal{O}$: in particular, we exhibit six infinite families of congruent blocks whose corresponding posets of highest weights consist of the partitions in question. These posets, in turn, lead to generalizations of the Littlewood identities and their corresponding BGG complexes. Our results in this paper shed light on the surprising combinatorics underlying the work of Enright and Willenbring (2004).