DIAS: Distance-based Attention Sparsity for Ultra-Long-Sequence Transformer with Tree-like Processing-in-Memory Architecture.
Zekai Chen, Yiming Chen, Teng Wan, Tianyi Yu, Yu Wang, Huazhong Yang, Xueqing Li
Browse the full DAC paper archive.
Zekai Chen, Yiming Chen, Teng Wan, Tianyi Yu, Yu Wang, Huazhong Yang, Xueqing Li
Browse the full DAC paper archive.