the output matrix is a square matrix and infers the matrix size from k and the
innermost dimension of `diagonal`.}]>:$num_rows,
    Arg<TF_Int32Tensor, [{The number of columns of the output matrix. If it is not provided, the op
assumes the output matrix is a square matrix and infers the matrix size from
k and the innermost dimension of `diagonal`.}]>:$num_cols,

    memory usage.
*   Fixed bfloat16 integration of LIBXSMM sparse mat-mul.
*   Improved performance and reduce memory usage by allowing ops to forward
    input buffers to output buffers and perform computations in-place.
*   Improved the performance of CPU assignment for strings.
*   Speed up matrix * vector multiplication and matrix * matrix with unknown
    shapes.