Note: You can now directly parse TAG grammars in the tulipac format with Alto. This will probably be faster than parsing with TuLiPA directly. Tulipac itself is no longer under development.
TuLiPA is the Tübingen Parsing Architecture. It is an excellent parser for teaching tree-adjoining grammar, except that it assumes an XML format for TAG grammars that is really complicated to work with.
Tulipac, the TuLiPA grammar compiler, fixes this problem by providing a front-end that compiles TAG grammars in a human-readable format to the XML format assumed by TuLiPA.