@inproceedings{liao-etal-2026-zipage,
    title = "Zipage: Maintain High Request Concurrency for {LLM} Reasoning through Compressed {P}aged{A}ttention",
    author = "Liao, Mengqi  and
      Wang, Lu  and
      Zhang, Chaoyun  and
      Qiao, Bo  and
      Qin, Si  and
      Lin, Qingwei  and
      Rajmohan, Saravan  and
      Zhang, Dongmei  and
      Wan, Huaiyu",
    editor = "Liakata, Maria  and
      Moreira, Viviane P.  and
      Zhang, Jiajun  and
      Jurgens, David",
    booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {ACL} 2026",
    month = jul,
    year = "2026",
    address = "San Diego, California, United States",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-acl/2026.findings-acl.381/",
    pages = "7716--7737",
    ISBN = "979-8-89176-395-1",
    abstract = "With reasoning becoming the generative paradigm for large language models, the memory bottleneck caused by KV cache during the inference phase has become a critical factor limiting high-concurrency service capabilities. Although existing KV cache eviction methods address the memory issue, most of them are impractical for industrial-grade applications. This paper introduces Compressed PagedAttention, a method that combines token-wise KV cache eviction with PagedAttention. We propose a comprehensive scheduling strategy and support prefix caching and asynchronous compression for Compressed PagedAttention. Based on this, we have developed a high-concurrency inference engine, Zipage. On large-scale mathematical reasoning tasks, Zipage achieves around 95{\%} of the performance of Full KV inference engines while delivering over 2.1 speedup."
}