High-order Semantic Role Labeling

Findings of the Association for Computational Linguistics: EMNLP 2020, pages 1134–1151 November 16 - 20, 2020. c⃝2020 Association for Computational Linguistics 1134 High-order Semantic Role Labeling Zuchao Li1,2,3, Hai Zhao1,2,3,∗, Rui Wang4, and Kevin Parnow1,2,3 1Department of Computer Science and Engineering, Shanghai Jiao Tong University (SJTU) 2Key Laboratory of Shanghai Education Commission for Intelligent Interaction and Cognitive Engineering, Shanghai Jiao Tong University, Shanghai, China 3MoE Key Lab of Artificial Intelligence, AI Institude, Shanghai Jiao Tong University, China 4National Institute of Information and Communications Technology (NICT), Kyoto, Japan charlee@sjtu.edu.cn, zhaohai@cs,sjtu.edu.cn, wangrui@nict.go.jp, parnow@sjtu.edu.cn Abstract Semantic role labeling is primarily used to identify predicates, arguments, and their se- mantic relationships. Due to the limitations of modeling methods and the conditions of pre-identified predicates, previous work has focused on the relationships between predi- cates and arguments and the correlations be- tween arguments at most, while the correla- tions between predicates have been neglected for a long time. High-order features and struc- ture learning were very common in model- ing such correlations before the neural net- work era. In this paper, we introduce a high- order graph structure for the neural semantic role labeling model, which enables the model to explicitly consider not only the isolated predicate-argument pairs but also the inter- action between the predicate-argument pairs. Experimental results on 7 languages of the CoNLL-2009 benchmark show that the high- order structural learning techniques are benefi- cial to the strong performing SRL models and further boost our baseline to achieve new state- of-the-art results. 1 Introduction Linguistic parsing seeks the syntactic/semantic re- lationships between language units, such as words or spans (chunks, phrases, etc.). The algorithms usually use factored representations of graphs to accomplish the target: a set of nodes and relational ∗ Corresponding author. This paper was partially sup- ported by National Key Research and Development Pro- gram of China (No. 2017YFB0304100), Key Projects of National Natural Science Foundation of China (U1836222 and 61733011), Huawei-SJTU Long Term AI Project, Cutting- edge Machine Reading Comprehension and Language Model. Rui Wang was partially supported by JSPS grant-in-aid for early-career scientists (19K20354): “Unsupervised Neural Ma- chine Translation in Universal Scenarios” and NICT tenure- track researcher startup fund “Toward Intelligent Machine Translation”. arcs. The types of features that the model can ex- ploit in the inference depend on the information included in the factorized parts. Before the introduction of deep neural networks, in syntactic parsing (a kind of linguistic parsing), several works (McDonald and Pereira, 2006; Car- reras, 2007; Koo and Collins, 2010; Zhang and McDonald, 2012; Ma and Zhao, 2012) showed that high-order parsers utilizing richer factorization in- formation achieve higher accuracy than low-order ones due to the extensive decision history that can lead to significant improvements in inference (Chen et al., 2010). Semantic role labeling (SRL) (Gildea and Juraf- sky, 2002; Zhao and Kit, 2008; Zhao et al., 2009b, 2013) captures the predicate-argument structure of a given sentence, and it is defined as a shallow se- mantic parsing task, which is also a typical linguis- tic …

Content truncated to 500 words