Enabling Lie-Algebraic Classical Simulation beyond Free Fermions

Adelina Bärligea, Matthew L. Sims-Goh, Jakob S. Kottmann (Apr 21 2026).

Abstract: Efficient classical simulation has matured to a critical component of the quantum computing stack, driving hardware validation, algorithm design, and the study of structured quantum dynamics. Lie-algebraic simulation ($\mathfrak{g}$-sim) is a compelling approach: it replaces exponentially large Hilbert-space evolution by dynamics in a reduced adjoint space whose dimension is set by the dynamical Lie algebra (DLA) of the circuit, enabling efficient simulation whenever the DLA grows only polynomially with system size. In practice, however, existing applications of $\mathfrak{g}$-sim have been confined to free-fermionic (matchgate) regimes, and it has been unclear how to extend the paradigm to other structured circuits whose generators may have large Pauli expansions. Here we enable Lie-algebraic classical simulation beyond free fermions by identifying additional non-trivial families of polynomial-dimensional DLAs and introducing symmetry-adapted basis representations that make the adjoint space mapping tractable. In particular, we develop an explicit Pauli orbit basis for permutation-equivariant dynamics, supporting cubic-dimensional algebras despite exponential Pauli support, and a subspace-adapted (modified) generalized Gell-Mann basis for bounded Hamming-weight ($U(1)$-equivariant) dynamics, yielding polynomial costs on fixed excitation sectors. Together with streamlined routines for free-fermionic Pauli algebras and translation-invariant variants thereof, these constructions significantly broaden the practical scope of $\mathfrak{g}$-sim as a unifying simulation tool for structured quantum dynamics. Numerical benchmarks confirm favorable preprocessing scaling and validate large-scale proof-of-concept simulations.

Arxiv: https://arxiv.org/abs/2604.16701