exponent raction (10 bit) Question 5t The shortest IEEE standard for repre-. senting rational numbers is called half-precision floating point. It uses a 16-bit word partitioned as in the diagram at right. (This diagram is taken from the Wikipedia article on the subject, where more details can be found.) sign 5 bit) 15 10 As described in lectures, to store a rational number r it is first represented as (-1)"X m x 2n with 1 m