Look at your notes and/or text book for the definition of two point distribution.

I would assume that it is a probability distribution with all the probability mass at two points.

For example p(x)=0 for all x in R, except at x=0, where p(0)=1/2 and x=1, where p(1)=1/2.

RonL