@article{loopserveanadaptivedualphasellm,
  title = {LoopServe: An Adaptive Dual-phase LLM Inference Acceleration System for Multi-Turn Dialogues},
  author = {Haoyang Li and Zhanchao Xu and Yiming Li and Xuejia Chen and Darian Li and Anxin Tian and Qingfa Xiao and Cheng Deng and Jun Wang and Qing Li and Lei Chen and Mingxuan Yuan},
  year = {2025},
  eprint = {2507.13681},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2507.13681v1},
}