arXiv AI· Rana Aref Salama, Abdou Youssef, Mona Diab·· 2 天前评分40
WNet: Discrete Wavelets Transform for Efficient Token Mixing
摘要
原文摘录（自动中文摘要暂不可用）：WNet: Discrete Wavelets Transform for Efficient Token Mixing In a Transformer, token mixing is the step that lets each token draw information from other tokens, and it dominates the cost of encoding long sequences. Self-attention does this mixing very well: every token weighs every other token by content, which gives strong contextual modeling. That all-pair
推荐理由
自动中文摘要尚未通过校验，请核对原文与完整来源。
本站只提供摘要与原文入口。完整内容请阅读原文。
来源：arXiv AI · arxiv.org