What is the state of the art with respect to recognizing connotations in natural languages?
- Trump is a better president than Obama. [Praising]
- Trump is the worst president since Obama. [Insulting]
- The rock star did not infect over 100 groupies. [Defending against rumor]
- The rock star infected no more than 100 groupies. [Attacking (0 is no more than 100)]
In each example, both statements logically mean exactly the same thing, but any human hearing them would interpret them as having quite opposite meanings.
How well can current natural language processors recognize the difference between logically equivalent statements?