Abstract: Numerous studies have proposed hardware architectures to accelerate sparse matrix multiplication, but these approaches often incur substantial area and power overhead, significantly ...