1
Higher-order Linear Attention
突破二次复杂度瓶颈,高阶线性注意力HLA以线性时间实现更强交互,为长上下文模型提供新方向。
arXiv:2510.27258v3 Announce Type: replace-cross Abstract: The quadratic cost of scaled dot-product attention is a central obstacle to scaling autoregr…
突破二次复杂度瓶颈,高阶线性注意力HLA以线性时间实现更强交互,为长上下文模型提供新方向。
arXiv:2510.27258v3 Announce Type: replace-cross Abstract: The quadratic cost of scaled dot-product attention is a central obstacle to scaling autoregr…