Syntax-aware Multilingual Semantic Role Labeling

Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing, pages 5350–5359, Hong Kong, China, November 3–7, 2019. c⃝2019 Association for Computational Linguistics 5350 Syntax-aware Multilingual Semantic Role Labeling Shexia He1,2,3,∗, Zuchao Li1,2,3,∗, Hai Zhao1,2,3,† 1Department of Computer Science and Engineering, Shanghai Jiao Tong University 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 Institute, Shanghai Jiao Tong University {heshexia, charlee}@sjtu.edu.cn, zhaohai@cs.sjtu.edu.cn Abstract Recently, semantic role labeling (SRL) has earned a series of success with even higher performance improvements, which can be mainly attributed to syntactic integration and enhanced word representation. However, most of these efforts focus on English, while SRL on multiple languages more than English has received relatively little attention so that is kept underdevelopment. Thus this paper in- tends to fill the gap on multilingual SRL with special focus on the impact of syntax and con- textualized word representation. Unlike exist- ing work, we propose a novel method guided by syntactic rule to prune arguments, which enables us to integrate syntax into multilingual SRL model simply and effectively. We present a unified SRL model designed for multiple languages together with the proposed uniform syntax enhancement. Our model achieves new state-of-the-art results on the CoNLL- 2009 benchmarks of all seven languages. Be- sides, we pose a discussion on the syntactic role among different languages and verify the effectiveness of deep enhanced representation for multilingual SRL. 1 Introduction Semantic role labeling (SRL) aims to derive the meaning representation such as an instantiated predicate-argument structure for a sentence. The currently popular formalisms to represent the se- mantic predicate-argument structure are based on dependencies and spans. Their main difference is that dependency SRL annotates the syntac- tic head of argument rather than the entire con- stituent (span), and this paper will focus on the ∗ These authors made equal contribution.† Correspond- ing author. This paper was partially supported by Na- tional Key Research and Development Program of China (No. 2017YFB0304100) and Key Projects of National Nat- ural Science Foundation of China (No. U1836222 and No. 61733011). dependency-based SRL. Be it dependency or span, SRL plays a critical role in many natural language processing (NLP) tasks, including information ex- traction (Christensen et al., 2011), machine trans- lation (Xiong et al., 2012) and question answering (Yih et al., 2016). Almost all of traditional SRL methods relied heavily on syntactic features, which suffered the risk of erroneous syntactic input, leading to unde- sired error propagation. To alleviate this inconve- nience, researchers as early as Zhou and Xu (2015) propose neural SRL models without syntactic in- put. Cai et al. (2018) employ the biaffine atten- tional mechanism (Dozat and Manning, 2017) for dependency-based SRL. In the meantime, a series of studies (Roth and Lapata, 2016; Marcheggiani and Titov, 2017; Strubell et al., 2018; Li et al., 2018) have introduced syntactic clue in creative ways for further performance …

Content truncated to 500 words