Do dependency parsing metrics correlate with human judgments?

Barbara Plank, H{\'e}ctor Mart{\'\i}nez Alonso, {\v{Z}}eljko Agi{\'c}, Danijela Merkler, Anders S{\o}gaard