“…The combination of high parsing speed of transition systems with the accuracy of the attached statistical models have paved the way for practical applications of parsing and similar data transformations enabled by these systems (Yamada and Matsumoto, 2003;Nivre and Scholz, 2004;Zhang and Nivre, 2011;Chen and Manning, 2014;Dyer et al, 2015;Andor et al, 2016;Kiperwasser and Goldberg, 2016;Shi et al, 2017). Transition systems may also be used to encode dependency trees, DAGs and other ordered digraphs, and to connect these to the classical formal language theory and to the problems of graph representation (Turán, 1984;Farzan and Munro, 2013;Yli-Jyrä, 2019), graph enumeration (Pólya, 1937;Conte et al, 2018;Yli-Jyrä, 2019), integer sequence discovery (Hoppe and Petrone, 2016;Yli-Jyrä and Gómez-Rodríguez, 2017), maximum subgraph inference (Conte et al, 2019;Yli-Jyrä and Gómez-Rodríguez, 2017), algebraic representations of graph queries (Courcelle, 1990;Ogawa, 2004;Yli-Jyrä and Gómez-Rodríguez, 2017), encoder-decoder parsing (Vinyals et al, 2015;Strzyz et al, 2019) and parsing as sequence labeling .…”