Argument Pair Extraction with Mutual Guidance and Inter-sentence Relation Graph

Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pages 3923–3934 November 7–11, 2021. c⃝2021 Association for Computational Linguistics 3923 Argument Pair Extraction with Mutual Guidance and Inter-sentence Relation Graph Jianzhu Bao1,2, Bin Liang1,2, Jingyi Sun1,2, Yice Zhang1,2 Min Yang3, Ruifeng Xu1,4∗ 1Harbin Institute of Technology (Shenzhen), China 2Joint Lab of China Merchants Securities and HITSZ 3Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences 4Peng Cheng Laboratory, Shenzhen, China jianzhubao@gmail.com, bin.liang@stu.hit.edu.cn sunjingyihit@gmail.com, zhangyc_hit@163.com min.yang@siat.ac.cn, xuruifeng@hit.edu.cn Abstract Argument pair extraction (APE) aims to ex- tract interactive argument pairs from two pas- sages of a discussion. Previous work stud- ied this task in the context of peer review and rebuttal, and decomposed it into a sequence labeling task and a sentence relation classi- fication task. However, despite the promis- ing performance, such an approach obtains the argument pairs implicitly by the two decom- posed tasks, lacking explicitly modeling of the argument-level interactions between argu- ment pairs. In this paper, we tackle the APE task by a mutual guidance framework, which could utilize the information of an argument in one passage to guide the identification of ar- guments that can form pairs with it in another passage. In this manner, two passages can mu- tually guide each other in the process of APE. Furthermore, we propose an inter-sentence re- lation graph to effectively model the inter- relations between two sentences and thus fa- cilitates the extraction of argument pairs. Our proposed method can better represent the holis- tic argument-level semantics and thus explic- itly capture the complex correlations between argument pairs. Experimental results show that our approach significantly outperforms the current state-of-the-art model. 1 Introduction Argumentation mining has received increasing re- search attention in recent years. Existing studies can be categorized into monological argumentation (Stab and Gurevych, 2014; Eger et al., 2017; Potash et al., 2017; Kuribayashi et al., 2019) and dialog- ical argumentation (Swanson et al., 2015; Morio and Fujita, 2018; Chakrabarty et al., 2019), with the former identifying the argumentation structure of a single monological document, and the latter fo- ∗Corresponding Author cusing on the analysis of argumentation in debates or discussions. Argument pair extraction (APE) is a new task within the field of dialogical argumentation, aim- ing at extracting interactive argument pairs from two argumentative passages of a discussion. Cheng et al. (2020) investigated this task in the context of peer review and rebuttal, as they involve rich argumentative and interactive discussions. An ex- ample of APE is shown in Figure 1, where a review passage and its corresponding rebuttal passage are segmented into arguments and non-arguments at sentence level. The arguments in review can form argument pairs with the arguments in rebuttal, ac- cording to the points they discuss. APE is a highly challenging task because we need to understand not only the argumentation structure presented by each side of the discussion, but also the interaction of arguments between the participants. The interactions between arguments can be complicated, for example, one argument may be paired …

Content truncated to 500 words