Learning Advanced Self-Attention for Linear Transformers in the Singular Value Domain

Re-assign community