@article{allocmoebudgetawareexpertactivationalloc,
  title = {Alloc-MoE: Budget-Aware Expert Activation Allocation for Efficient Mixture-of-Experts Inference},
  author = {Baihui Liu and Kaiyuan Tian and Wei Wang and Zhaoning Zhang and Linbo Qiao and Dongsheng Li},
  year = {2026},
  eprint = {2604.08133},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2604.08133},
}