Extractive Summarization Considering Discourse and Coreference Relations based on Heterogeneous Graph

Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics, pages 3046–3052 April 19 - 23, 2021. ©2021 Association for Computational Linguistics 3046 Extractive Summarization Considering Discourse and Coreference Relations based on Heterogeneous Graph Yin Jou Huang Kyoto University huang@nlp.ist.i.kyoto-u.ac.jp Sadao Kurohashi Kyoto University kuro@i.kyoto-u.ac.jp Abstract Modeling the relations between text spans in a document is a crucial yet challenging problem for extractive summarization. Various kinds of relations exist among text spans of differ- ent granularity, such as discourse relations be- tween elementary discourse units and corefer- ence relations between phrase mentions. In this paper, we propose a heterogeneous graph based model for extractive summarization that incorporates both discourse and coreference relations. The heterogeneous graph contains three types of nodes, each corresponds to text spans of different granularity. Experi- mental results on a benchmark summariza- tion dataset verify the effectiveness of our pro- posed method. 1 Introduction Automatic summarization aims to condense the information of the input document into a shorter summary. The task has two main paradigms: ex- tractive summarization and abstractive summariza- tion. Generating summary sentences from scratch, abstractive summarizers can generate concise and flexible summaries. However, they also suffer from the problem of not being able to reproduce fac- tual details correctly (See et al., 2017). On the other hand, extractive summarization aims to select salient text spans (mostly sentences) from the input document. Compared to abstractive summarizers, extractive summarizers have the advantage of being efficient and factually reliable. In this paper, we will focus on extractive summarization. For extractive summarization, it is crucial to model the relations between text spans through- out the document. Between text spans of different granularity, there exist many different kinds of rela- tions (Figure 1). For example, coreference relations exist between mention phrases of the same entity, and discourse relations exist between Elementary Discourse Units (EDUs) within a document. Due Figure 1: Relations among text spans of different gran- ularity. to its complex nature, modeling the various rela- tions among text spans of a document remains an open challenge. To capture inter-sentential relations, some recent works utilize recurrent neural networks (RNNs) or Transformer (Vaswani et al., 2017) based encoders on top of the acquired sentence representations (Cheng and Lapata, 2016; Nallapati et al., 2016; Liu and Lapata, 2019). However, empirical obser- vations show that these sentence-level encoders do not bring much performance gain (Liu and Lap- ata, 2019). Graph structure is an intuitive way to model long-range dependencies among text spans throughout a document. Early works build connec- tivity graphs based on content similarity between sentences (Erkan and Radev, 2004; Mihalcea and Tarau, 2004). Some recent works incorporate dis- course or coreference relations into the graph struc- ture and utilize graph neural networks (GNNs) to obtain a high-level representation of text spans (Ya- sunaga et al., 2017; Xu and Durrett, 2019; Xu et al., 2020). Most of these works operate on homoge- neous graphs with only one type of nodes, such as Approximate Discourse Graph (ADG) …

Content truncated to 500 words