Abstract:Extractive document summarization is a fundamental task in natural language processing (NLP). Recently, several Graph Neural Networks (GNNs) are proposed for this task. However, most existing GNN-based models can neither effectively encode semantic nodes of multiple granularity level apart from sentences nor substantially capture different cross-sentence meta-paths. To address these issues, we propose MHgatSum, a novel Multi-granularity Heterogeneous Graph ATtention networks for extractive document SUMmarization. Specifically, we first build a multi-granularity heterogeneous graph (HetG) for each document, which is better to represent the semantic meaning of the document. The HetG contains not only sentence nodes but also multiple other granularity effective semantic units with different semantic levels, including keyphrases and topics. These additional nodes act as the intermediary between sentences to build the meta-paths involved in sentence node (i.e., Sentence-Keyphrase-Sentence and Sentence-Topic-Sentence). Then, we propose a heterogeneous graph attention networks to embed the constructed HetG for extractive summarization, which enjoys multi-granularity semantic representations. The model is based on a hierarchical attention mechanism, including node-level and semantic-level attentions. The node-level attention can learn the importance between a node and its meta-path based neighbors, while the semantic-level attention is able to learn the importance of different meta-paths. Moreover, to better integrate sentence global knowledge, we further incorporate sentence node global importance in local node-level attention. We conduct empirical experiments on two benchmark datasets, which demonstrates the superiority of MHgatSum over previous SOTA models on the task of extractive summarization.

Heterogeneous graphormer for extractive multimodal summarization

HGTMFS: A Hypergraph Transformer Framework for Multimodal Summarization

An Unsupervised Video Summarization Method Based on Multimodal Representation.

Multimodal Summarization with Guidance of Multimodal Reference

Multi-View Metrics Enhanced Heterogeneous Graph Neural Network for Extractive Summarization

MHMS: Multimodal Hierarchical Multimedia Summarization

Abstractive Sentence Summarization with Guidance of Selective Multimodal Reference.

Hierarchical Cross-Modality Semantic Correlation Learning Model for Multimodal Summarization

Self-supervised opinion summarization with multi-modal knowledge graph

SgSum: Transforming Multi-document Summarization into Sub-graph Selection

Multi-task Hierarchical Heterogeneous Fusion Framework for multimodal summarization

Multi-granularity heterogeneous graph attention networks for extractive document summarization

A Multi-Granularity Heterogeneous Graph for Extractive Text Summarization

Leveraging Entity Information for Cross-Modality Correlation Learning: The Entity-Guided Multimodal Summarization

Multiplex Graph Neural Network for Extractive Text Summarization

Heterogeneous Graph Neural Networks for Extractive Document Summarization

SITransformer: Shared Information-Guided Transformer for Extreme Multimodal Summarization

VMSMO: Learning to Generate Multimodal Summary for Video-based News Articles

Unsupervised Extractive Summarization with Heterogeneous Graph Embeddings for Chinese Document

Semantics-Consistent Cross-domain Summarization via Optimal Transport Alignment

Multimodal Graph for Unaligned Multimodal Sequence Analysis via Graph Convolution and Graph Pooling