Current version of the annotator uses the output of the MATE parser (https://code.google.com/archive/p/mate-tools/wikis/ParserAndModels.wiki) in the ConLL format. For English, a joint model for tagging and parsing is used (https://drive.google.com/file/d/0B-qbj-8rtoUMWWlINVhqdTU0bjQ/view?usp=sharing). For German and French, the data first needs to be lemmatized (DE: https://drive.google.com/file/d/0B-qbj-8rtoUMaUVsWUFuOE81ZW8/edit?usp=sharing, FR: https://drive.google.com/file/d/0B-qbj-8rtoUMMEYwY0FFLUVmeEU/edit?usp=sharing) and then parsed using the joint models for tagging, morphology and parsing (DE: https://drive.google.com/file/d/0B-qbj-8rtoUMLUg5NGpBVW9JNkE/edit?usp=sharing, FR: https://drive.google.com/file/d/0B-qbj-8rtoUMYV8yalAtcFk0WFE/edit?usp=sharing).