@article{differentiallyprivateandcommunicationeff,
  title = {Differentially Private and Communication Efficient Large Language Model Split Inference via Stochastic Quantization and Soft Prompt},
  author = {Yujie Gu and Richeng Jin and Xiaoyu Ji and Yier Jin and Wenyuan Xu},
  year = {2026},
  eprint = {2602.11513},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2602.11513},
}