@article{puzzlemoeefficientcompressionoflargemixt,
  title = {PuzzleMoE: Efficient Compression of Large Mixture-of-Experts Models via Sparse Expert Merging and Bit-packed inference},
  author = {Yushu Zhao and Zheng Wang and Minjia Zhang},
  year = {2025},
  eprint = {2511.04805},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2511.04805},
}