Clarifications and Optimizations on Rounding for IEEE-compliant Floating-Point Multiplication | IEEE Conference Publication | IEEE Xplore