Local Ai
FlashCart: Fast Cartesian Tensor Products for Equivariant Interatomic Potentials
arXiv:2610.06409v1 Announce Type: new Abstract: Machine-learned interatomic potentials extend atomistic simulations beyond the length- and timescales accessible to electronic-structure methods. Howeve
arXiv:2610.06409v1 Announce Type: new Abstract: Machine-learned interatomic potentials extend atomistic simulations beyond the length- and timescales accessible to electronic-structure methods. However, the computational cost of equivariant architectures limits the local correlations they can represent in practice and therefore their achievable accuracy. Here we introduce FlashCart, which makes higher-order correlations affordable by combining generated GPU kernels with an architecture that recursively builds equivariant features and compresses them to a fixed width at each step. We express tensor products in independent Cartesian components and symbolically simplify them and their derivatives, producing fused kernels that often outperform optimized spherical counterparts. We then show that increasing correlation order improves accuracy more efficiently than increasing width, depth, or tensor rank. On SPICE-MACE-OFF, FlashCart models advance the measured accuracy-efficiency frontier: a model with 5.6 million parameters achieves lower energy and force errors and 10imes faster inference than a transformer with 189 million parameters.
Related
- Equivariant Cellular Sheaves for Molecular Electronic Structure: Bridging Sheaf Cohomology and E(3)-Equivariant Hamiltonian Learning
- Electronic Density versus Geometry for Machine-Learned Molecular Absorption Spectra
- Transformers Discover Molecular Structure Without Graph Priors
Source: arXiv cs.LG | 2026-10-06