Deep Learning Galerkin Transformer: a linear attention without softmax Galerkin Transformer: a linear attention without softmax 03 October 2021