I posted links to some existing parsers on the task homepage
http://pete.yuret.com. You do not have to use any of these parsers,
you do not even have to use a parser at all -- out of the box
approaches are highly encouraged. However, to get a quick baseline
system using an existing parser may be a good way to start.
I also posted a script, conll-entailments.pl, that illustrates how
short entailments may be automatically generated from a dependency
parse. Does not work well but may give you some ideas.