A Unified Generative Framework for Aspect-based Sentiment Analysis

Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing, pages 210–219 August 1–6, 2021. ©2021 Association for Computational Linguistics 210 Capturing Event Argument Interaction via A Bi-Directional Entity-Level Recurrent Decoder Xi Xiangyu1,2, Wei Ye1,†, Shikun Zhang1,†, Quanxiu Wang3, Huixing Jiang2, Wei Wu2 1 National Engineering Research Center for Software Engineering, Peking University, Beijing, China 2 Meituan Group, Beijing, China 3 RICH AI, Beijing, China {xixy,wye,zhangsk}@pku.edu.cn Abstract Capturing interactions among event arguments is an essential step towards robust event argu- ment extraction (EAE). However, existing ef- forts in this direction suffer from two limita- tions: 1) The argument role type information of contextual entities is mainly utilized as train- ing signals, ignoring the potential merits of directly adopting it as semantically rich input features; 2) The argument-level sequential se- mantics, which implies the overall distribution pattern of argument roles over an event men- tion, is not well characterized. To tackle the above two bottlenecks, we formalize EAE as a Seq2Seq-like learning problem for the first time, where a sentence with a specific event trigger is mapped to a sequence of event ar- gument roles. A neural architecture with a novel Bi-directional Entity-level Recurrent De- coder (BERD) is proposed to generate argu- ment roles by incorporating contextual enti- ties’ argument role predictions, like a word- by-word text generation process, thereby dis- tinguishing implicit argument distribution pat- terns within an event more accurately. 1 Introduction Event argument extraction (EAE), which aims to identify the entities serving as event arguments and classify the roles they play in an event, is a key step towards event extraction (EE). For example, given that the word “fired” triggers an Attack event in the sentence “In Baghdad, a cameraman died when an American tank fired on the Palestine Hotel” , EAE need to identify that “Baghdad”, “camera- man”, “American tank”, and “Palestine hotel” are arguments with Place, Target, Instrument, and Tar- get as roles respectively. Recently, deep learning models have been widely applied to event argument extraction and †Corresponding authors. achieved significant progress(Chen et al., 2015; Nguyen et al., 2016; Sha et al., 2018; Yang et al., 2019; Wang et al., 2019b; Zhang et al., 2020; Du and Cardie, 2020). Many efforts have been devoted to improving EAE by better characterizing argu- ment interaction, categorized into two paradigms. The first one, named inter-event argument inter- action in this paper, concentrates on mining infor- mation of the target entity (candidate argument) in the context of other event instances (Yu et al., 2011; Nguyen et al., 2016), e.g., the evidence that a Victim argument for the Die event is often the Target argument for the Attack event in the same sentence. The second one is intra-event argu- ment interaction, which exploits the relationship of the target entity with others in the same event instance (Yu et al., 2011; Sha et al., 2016, 2018). We focus on the second paradigm in this paper. Despite their promising results, existing …

Content truncated to 500 words