Abstract:Let $G$ be a complex reductive algebraic group. In <a class="link-https" data-arxiv-id="2108.03453" href="https://arxiv.org/abs/2108.03453">arXiv:2108.03453</a> Ivan Losev, Lucas mason-Brown and the third-named author suggested a symplectic duality between nilpotent Slodowy slices in $\mathfrak{g}^\vee$ and affinizations of certain $G$-equivariant covers of special nilpotent orbits. In this paper, we study the various versions of Hikita conjecture for this pair. We show that the original statement of the conjecture does not hold for the pairs in question and propose a refined version. We discuss the general approach towards the proof of the refined Hikita conjecture and prove this refined version for the parabolic Slodowy varieties, which includes many of the cases considered in <a class="link-https" data-arxiv-id="2108.03453" href="https://arxiv.org/abs/2108.03453">arXiv:2108.03453</a> and more. Applied to the setting of <a class="link-https" data-arxiv-id="2108.03453" href="https://arxiv.org/abs/2108.03453">arXiv:2108.03453</a>, the refined Hikita conjecture explains the importance of special unipotent ideals from the symplectic duality point of view.
We also discuss applications of our results. In the appendices, we discuss some classical questions in Lie theory that relate the refined version and the original version. We also explain how one can use our results to simplify some proofs of known results in the literature. As a combinatorial application of our results we observe an interesting relation between the geometry of Springer fibers and left Kazhdan-Lusztig cells in the corresponding Weyl group.
What problem does this paper attempt to address?
The main problem that this paper attempts to solve is about the validity of the Hikita - Nakajima conjecture in certain specific situations and the proposal and proof of its revised version. Specifically:
1. **Validity of the original Hikita - Nakajima conjecture**: The author first studies whether the original Hikita - Nakajima conjecture holds for a specific symmetric dual pair (i.e., \(X_ = \text{Spec}(C[[S(p_,q_)]]), X = \text{Spec}(C[[S(q,p)]])\)). By constructing specific counter - examples, they find that the original conjecture is not always correct in this case.
2. **Revised Hikita - Nakajima conjecture**: To make up for the deficiencies of the original conjecture, the author proposes a revised version of the Hikita - Nakajima conjecture. This revised version mainly involves two aspects:
- **Revision on the homological side**: On the homological side, the author considers the image of \(H^*_{ZL_ \times \mathbb{C}^*}(S(p_,q_))\), instead of directly using \(H^*_{ZL_ \times \mathbb{C}^*}(S(p_,q_))\).
- **Revision on the quantum side**: On the quantum side, the author considers the image of \(C_\nu(A_{\hbar, X(l)}(S(q,p)))\), instead of directly using \(C_\nu(A_{\hbar, X(l)}(S(q,p)))\).
3. **Main results**: The author proves that the revised Hikita - Nakajima conjecture holds for parabolic Slodowy varieties. Specifically, they prove the following theorem:
\[
\text{Im}\left(H^*_{ZL_ \times \mathbb{C}^*}(T^*Q) \to H^*_{ZL_ \times \mathbb{C}^*}(S(p_,q_))\right) \cong \text{Im}\left(C_\nu(A_{\hbar, X(l)}(T^*B)) \to C_\nu(A_{\hbar, X(l)}(S(q,p))) \to \Gamma(S(q,p)^{\nu(\mathbb{C}^*)}, C_\nu(D_{X(l),\hbar}(S(q,p))))\right)
\]
4. **Applications and motivations**:
- **Connection with one - parameter characters**: The revised Hikita - Nakajima conjecture explains the appearance of special one - parameter central characters in the canonical quantization of orbit covers.
- **Springer fibers and Kazhdan - Lusztig cells**: The revised Hikita - Nakajima conjecture provides a new bridge connecting representation theory and geometry, especially the relationship between the set - theoretic support of Springer fibers and the Kazhdan - Lusztig cell theory.
Through these works, the author not only solves the inapplicability of the original Hikita - Nakajima conjecture in some cases but also proposes and proves a more precise revised version, providing new tools and perspectives for further research in related fields.