📝 SummarXiv

Abstract

We develop LREI (Low-Rank Eigenmode Integration), a memory- and time-efficient scheme for solving quantum Landau-Lifshitz (q-LL) and quantum Landau-Lifshitz-Gilbert (q-LLG) equations, which govern spin dynamics in open quantum systems. Although system size grows exponentially with the number of spins, our approach exploits the low-rank structure of the density matrix and the sparsity of Hamiltonians to avoid full matrix computations. By representing density matrices via low-rank factors and applying Krylov subspace methods for partial eigendecompositions, we reduce the per-step complexity of Runge-Kutta and Adams-Bashforth schemes from $\mathcal{O}(N^3)$ to $\mathcal{O}(r^2N)$, where $N = 2^n$ is the Hilbert space dimension for $n$ spins and $r \ll N$ the effective rank. Similarly, memory costs shrink from $\mathcal{O}(N^2)$ to $\mathcal{O}(rN)$, since no full $N\times N$ matrices are formed. A key advance is handling the invariant subspace of zero eigenvalues. By using Householder reflectors built for the dominant eigenspace, we perform the solution entirely without large matrices. For example, a time step of a twenty-spin system, with density matrix size over one million, now takes only seconds on a standard laptop. Both Runge-Kutta and Adams-Bashforth methods are reformulated to preserve physical properties of the density matrix throughout evolution. This low-rank algorithm enables simulations of much larger spin systems, which were previously infeasible, providing a powerful tool for comparing q-LL and q-LLG dynamics, testing each model validity, and probing how quantum features such as correlations and entanglement evolve across different regimes of system size and damping.