@article{250510860,
  title = {On DeepSeekMoE: Statistical Benefits of Shared Experts and Normalized Sigmoid Gating},
  author = {Huy Nguyen and Thong T. Doan and Quang Pham and Nghi D. Q. Bui and Nhat Ho and Alessandro Rinaldo},
  year = {2025},
  eprint = {2505.10860},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2505.10860v1},
}