PAT: Accelerating LLM Decoding via Prefix-Aware Attention with Resource Efficient Multi-Tile Kernel.

Jinjun Yi, Zhixin Zhao, Yitao Hu, Ke Yan, Weiwei Sun, Hao Wang, Laiping Zhao, Yuhao Zhang, Wenxin Li, Keqiu Li

Browse the full ASPLOS paper archive.