11 字
1 分鐘
Transformers are RNNs: Fast Autoregressive Transformers with Linear Attention
Angelos Katharopoulos, Apoorv Vyas, Nikolaos Pappas, François Fleuret
Paper link: https://arxiv.org/abs/2006.16236
Transformers are RNNs: Fast Autoregressive Transformers with Linear Attention
https://www.poyu39.tw/posts/seminar/arxiv200616236/