@article{fastersmallerandsmartertaskawareexpertme,
  title = {Faster, Smaller, and Smarter: Task-Aware Expert Merging for Online MoE Inference},
  author = {Ziyi Han and Xutong Liu and Ruiting Zhou and Xiangxiang Dai and John C. S. Lui},
  year = {2025},
  eprint = {2509.19781},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2509.19781},
}