Universal Dependencies (UD) is a framework for cross-linguistically consistent treebank annotation that has so far been applied to over 50 languages (http://universaldependencies.org/). The framework is aiming to capture similarities as well as idiosyncrasies among typologically different languages (e.g., morphologically rich languages, pro-drop languages, and languages featuring clitic doubling). The goal in developing UD was not only to support comparative evaluation and cross-lingual learning but also to facilitate multilingual natural language processing and enable comparative linguistic studies.
After a period of rapid growth since the release of the first guidelines in October 2014 and the release of the second version of the guidelines in December 2016, we felt it was time to take stock and reflect on the theory and practice of UD, its use in research and development, and its future goals and challenges. We are returning to Gothenburg where UD, in its actual implementation, was born in the spring of 2014.
We received 29 submissions of which 24 were accepted. Submissions covered several topics: the workshop feature papers describing treebank conversion or creation, while others focus on resources useful for annotation; some work targets specific syntactic constructions and which analysis to adopt, sometimes with critiques of the choices made in UD; some papers exploit UD resources for parsing or downstream tasks, often in a cross-lingual setting, and others discuss the relation of UD to different frameworks.
We are honored to have two invited speakers: Mirella Lapata (School of Informatics, University of Edinburgh, Scotland), with a talk on “Universal Semantic Parsing”, and William Croft (Department of Linguistics, University of New Mexico, USA), speaking about “Using Typology to Develop Guidelines for Universal Dependencies”. Our invited speakers’ work target different aspects of UD: Mirella Lapata’s talk is an instance of how UD facilitates building downstream applications which can operate multilingually, whereas William Croft will address how UD and typological universals intersect.
We are grateful to the program committee, who worked hard and on a tight schedule to review the submissions and provided authors with valuable feedback. We thank Google, Inc. for its sponsorship which made it ossible to feature two invited talks. We also want to thank the organizing committee for their help; in particular Francis Tyers and Sebastian Schuster for their invaluable help with the conference software and these proceedings, as well as Sampo Pyysalo for setting up the website and providing immediate response for updating it.
We tried to set up the program to favor discussions, and we wish all participants a productive workshop!
Marie-Catherine de Marneffe and Joakim Nivre