[48] SAC: Accelerating and Structuring Self-Attention via Sparse Adaptive Connection

August 9, 2022 ยท 1 min ยท long8v ยท 

[21] cosFormer: Rethinking Softmax in Attention

April 20, 2022 ยท 3 min ยท long8v ยท 

[14] Longformer: The Long-Document Transformer

February 22, 2022 ยท 2 min ยท long8v ยท