Accelerating sparse matrix-vector multiplication on GPUs using bit-representation-optimized schemes | IEEE Conference Publication | IEEE Xplore