SemEval-2019 Task 1: Cross-lingual Semantic Parsing with UCCA

SemEval-2019 Task 1: Cross-lingual Semantic Parsing with UCCA Daniel Hershcovich Elior Sulem Zohar Aizenbud Ari Rappoport School of Computer Science and Engineering Hebrew University of Jerusalem {danielh,zohara,borgr,eliors,arir,oabend}@cs.huji.ac.il Leshem Choshen Omri Abend Abstract We present the SemEval 2019 shared task on Universal Conceptual Cognitive Annota- tion (UCCA) parsing in English, German and French, and discuss the participating systems and results. UCCA is a cross- linguistically applicable framework for se- mantic representation, which builds on exten- sive typological work and supports rapid an- notation. UCCA poses a challenge for ex- isting parsing techniques, as it exhibits reen- trancy (resulting in DAG structures), discon- tinuous structures and non-terminal nodes cor- responding to complex semantic units. The shared task has yielded improvements over the state-of-the-art baseline in all languages and settings. Full results can be found in the task’s website https://competitions. codalab.org/competitions/19160. 1 Overview Semantic representation is receiving growing at- tention in NLP in the past few years, and many proposals for semantic schemes have recently been put forth. Examples include Abstract Mean- ing Representation (AMR; Banarescu et al., 2013), Broad-coverage Semantic Dependencies (SDP; Oepen et al., 2016), Universal Decompositional Semantics (UDS; White et al., 2016), Parallel Meaning Bank (Abzianidze et al., 2017), and Uni- versal Conceptual Cognitive Annotation (UCCA; Abend and Rappoport, 2013). These advances in semantic representation, along with correspond- ing advances in semantic parsing, can potentially benefit essentially all text understanding tasks, and have already demonstrated applicability to a vari- ety of tasks, including summarization (Liu et al., 2015; Dohare and Karnick, 2017), paraphrase de- tection (Issa et al., 2018), and semantic evaluation (using UCCA; see below). In this shared task, we focus on UCCA parsing in multiple languages. After L graduation P H , U John A moved P to R Paris C A H A Figure 1: An example UCCA graph. One of our goals is to benefit semantic parsing in languages with less annotated resources by mak- ing use of data from more resource-rich languages. We refer to this approach as cross-lingual parsing, while other works (Zhang et al., 2017, 2018) de- fine cross-lingual parsing as the task of parsing text in one language to meaning representation in another language. In addition to its potential applicative value, work on semantic parsing poses interesting algo- rithmic and modeling challenges, which are often different from those tackled in syntactic parsing, including reentrancy (e.g., for sharing arguments across predicates), and the modeling of the inter- face with lexical semantics. UCCA is a cross-linguistically applicable se- mantic representation scheme, building on the established Basic Linguistic Theory typological framework (Dixon, 2010b,a, 2012). It has demon- strated applicability to multiple languages, includ- ing English, French and German, and pilot an- notation projects were conducted on a few lan- guages more. UCCA structures have been shown to be well-preserved in translation (Sulem et al., 2015), and to support rapid annotation by non- experts, assisted by an accessible annotation in- terface (Abend et al., 2017).1 UCCA has al- ready shown applicative value for text simplifica- 1https://github.com/omriabnd/UCCA-App …

Content truncated to 500 words