作者位次 #11/12ACL 2026Findings· Findings of ACL
Lil: Less is Less When Applying Post-Training Sparse-Attention Algorithms in Long-Decode Stage
Junhao Hu · Fangze Li · Mingtao Xu · Feifan Meng · Shiju Zhao · Tiancheng Hu · Ting Peng · Anmin Liu · Wenrui Huang · Chenxu Liu · Ziyue Hua · Tao Xie
摘要
Large language models (LLMs) demonstrate strong capabilities across a wide range of complex tasks and are increasingly deployed at scale, placing significant demands on inference efficiency. Prior work typically decomposes inference into prefill and decode stages, with the decode stage dominating...