@article{efficientmixtureofexpertsllminferencewit, title = {Efficient Mixture-of-Experts LLM Inference with Apple Silicon NPUs}, author = {Afsara Benazir and Felix Xiaozhu Lin}, year = {2026}, eprint = {2604.18788}, archivePrefix = {arXiv}, url = {https://arxiv.org/abs/2604.18788}, }