Each year the Conference on Computational Natural Language Learning (CoNLL) 1 features a shared task, in which participants train and test their systems on exactly the same data sets, in order to better compare systems. The tenth CoNLL (CoNLL-X) saw a shared task on Multilingual Dependency Parsing. In this paper, we describe how treebanks for 13 languages were converted into the same dependency format and how parsing performance was measured. We also give an overview of the parsing approaches that participants took and the results that they achieved. Finally, we try to draw general conclusions about multi-lingual parsing: What makes a particular language, treebank or annotation scheme easier or harder to parse and which phenomena are challenging for any dependency parser? Acknowledgement Many thanks to Amit Dubey and Yuval Krymolowski, the other two organizers of the shared task, for discussions, converting treebanks, writing software and helping with the papers. 2

We present a comparison between two systems for establishing syntactic and semantic dependencies: one that performs dependency parsing and semantic role labeling as a single task, and another that performs the two tasks in isolation. The systems are based on local memorybased classifiers predicting syntactic and semantic dependency relations between pairs of words. In a second global phase, the systems perform a deterministic ranking procedure in which the output of the local classifiers is combined per sentence into a dependency graph and semantic role labeling assignments for all predicates. The comparison shows that in the learning phase a joint approach produces better-scoring classifiers, while after the ranking phase the isolated approach produces the most accurate syntactic dependencies, while the joint approach yields the most accurate semantic role assignments.

3.1 Mbtg, the tagger generator................................. 4