SummaRuNNer: A Recurrent Neural Network based Sequence Model for Extractive Summarization of Documents

Abstract

We present SummaRuNNer, a Recurrent Neural Network (RNN) based sequence model for extractive summarization of documents and show that it achieves performance better than or comparable to state-of-the-art. Our model has the additional advantage of being very interpretable, since it allows visualization of its predictions broken up by abstract features such as information content, salience and novelty. Another novel contribution of our work is abstractive training of our extractive model that can train on human generated reference summaries alone, eliminating the need for sentence-level extractive labels.

Published at AAAI 2017, The Thirty-First AAAI Conference on Artificial Intelligence (AAAI-2017)

Cited by in corpus (103)

• HIBERT: Document Level Pre-training of Hierarchical Bidirectional Transformers for Document Summarization
• Graph-based Neural Multi-Document Summarization
• Text Summarization using Abstract Meaning Representation
• Efficient Adaptation of Pretrained Transformers for Abstractive Summarization
• Heterogeneous Graph Neural Networks for Extractive Document Summarization
• Deep Recurrent Generative Decoder for Abstractive Text Summarization
• Learning with fuzzy hypergraphs: a topical approach to query-oriented text summarization
• Investigating Entropy for Extractive Document Summarization
• QMSum: A New Benchmark for Query-based Multi-domain Meeting Summarization
• Sentence Centrality Revisited for Unsupervised Summarization
• Searching for Effective Neural Extractive Summarization: What Works and What's Next
• A Multi-task Learning Approach for Improving Product Title Compression with User Search Log Data
• Multi-modal Summarization for Video-containing Documents
• Attention Optimization for Abstractive Document Summarization
• A Survey on Recent Advances in Sequence Labeling from Deep Learning Models
• Extractive Summarization as Text Matching
• Systematically Exploring Redundancy Reduction in Summarizing Long Documents
• Generating Informative CVE Description From ExploitDB Posts by Extractive Summarization
• DSGPT: Domain-Specific Generative Pre-Training of Transformers for Text Generation in E-commerce Title and Review Summarization
• A Condense-then-Select Strategy for Text Summarization
• Biomedical text summarization using Conditional Generative Adversarial Network(CGAN)
• SEAL: Segment-wise Extractive-Abstractive Long-form Text Summarization
• What comes next? Extractive summarization by next-sentence prediction
• Massive Multi-Document Summarization of Product Reviews with Weak Supervision
• Continual BERT: Continual Learning for Adaptive Extractive Summarization of COVID-19 Literature
• Language Model as an Annotator: Exploring DialoGPT for Dialogue Summarization
• CO2Sum:Contrastive Learning for Factual-Consistent Abstractive Summarization
• Literature Retrieval for Precision Medicine with Neural Matching and Faceted Summarization
• Summary Level Training of Sentence Rewriting for Abstractive Summarization
• Pragmatically Informative Text Generation
• Reducing Quantity Hallucinations in Abstractive Summarization
• Multi-Modal Generative Adversarial Network for Short Product Title Generation in Mobile E-Commerce
• Neural Sentence Ordering Based on Constraint Graphs
• Topic-aware Pointer-Generator Networks for Summarizing Spoken Conversations
• A Closer Look at Data Bias in Neural Extractive Summarization Models
• Multi-document Summarization with Maximal Marginal Relevance-guided Reinforcement Learning
• Document Modeling with Graph Attention Networks for Multi-grained Machine Reading Comprehension
• LawSum: A weakly supervised approach for Indian Legal Document Summarization
• Self-Supervised Learning for Contextualized Extractive Summarization
• Answering while Summarizing: Multi-task Learning for Multi-hop QA with Evidence Extraction
• Unsupervised Extractive Summarization using Pointwise Mutual Information
• From Standard Summarization to New Tasks and Beyond: Summarization with Manifold Information
• An Integrated Approach for Keyphrase Generation via Exploring the Power of Retrieval and Extraction
• Question Answering as an Automatic Evaluation Metric for News Article Summarization
• Improving the Similarity Measure of Determinantal Point Processes for Extractive Multi-Document Summarization
• LeafNATS: An Open-Source Toolkit and Live Demo System for Neural Abstractive Text Summarization
• Joint Learning of Answer Selection and Answer Summary Generation in Community Question Answering
• The Style-Content Duality of Attractiveness: Learning to Write Eye-Catching Headlines via Disentanglement
• CSDS: A Fine-Grained Chinese Dataset for Customer Service Dialogue Summarization
• On Generating Extended Summaries of Long Documents
• A more abstractive summarization model
• Contextualized Rewriting for Text Summarization
• Abstractive Text Summarization by Incorporating Reader Comments
• Artificial Intelligence Driven UAV-NOMA-MEC in Next Generation Wireless Networks
• VT-SSum: A Benchmark Dataset for Video Transcript Segmentation and Summarization
• Topological Sort for Sentence Ordering
• PodSumm -- Podcast Audio Summarization
• Knowledge-guided Unsupervised Rhetorical Parsing for Text Summarization
• Towards Supervised Extractive Text Summarization via RNN-based Sequence Classification
• Discriminative Pre-training for Low Resource Title Compression in Conversational Grocery
• Multiple Instance Learning Networks for Fine-Grained Sentiment Analysis
• How well do you know your summarization datasets?
• An Effective Contextual Language Modeling Framework for Speech Summarization with Augmented Features
• LenAtten: An Effective Length Controlling Unit For Text Summarization
• Few-Shot Learning of an Interleaved Text Summarization Model by Pretraining with Synthetic Data
• Guiding Extractive Summarization with Question-Answering Rewards
• Specificity-Based Sentence Ordering for Multi-Document Extractive Risk Summarization
• Enhance Long Text Understanding via Distilled Gist Detector from Abstractive Summarization
• Fact-level Extractive Summarization with Hierarchical Graph Mask on BERT
• CNewSum: A Large-scale Chinese News Summarization Dataset with Human-annotated Adequacy and Deducibility Level
• CX DB8: A queryable extractive summarizer and semantic search engine
• Review-Driven Answer Generation for Product-Related Questions in E-Commerce
• Bringing Structure into Summaries: a Faceted Summarization Dataset for Long Scientific Documents
• Topic-Guided Abstractive Multi-Document Summarization
• Extractive Research Slide Generation Using Windowed Labeling Ranking
• Demoting the Lead Bias in News Summarization via Alternating Adversarial Learning
• Ranking sentences from product description & bullets for better search
• Abstractive and mixed summarization for long-single documents
• Extractive Summarizer for Scholarly Articles
• A Generative Approach to Titling and Clustering Wikipedia Sections
• SupMMD: A Sentence Importance Model for Extractive Summarization using Maximum Mean Discrepancy
• An Unsupervised Semantic Sentence Ranking Scheme for Text Documents
• Read, Highlight and Summarize: A Hierarchical Neural Semantic Encoder-based Approach
• Learning to Summarize Passages: Mining Passage-Summary Pairs from Wikipedia Revision Histories
• Generating summaries tailored to target characteristics
• Improving Abstractive Text Summarization with History Aggregation
• Attractive or Faithful? Popularity-Reinforced Learning for Inspired Headline Generation
• Using multiple ASR hypotheses to boost i18n NLU performance
• STRASS: A Light and Effective Method for Extractive Summarization Based on Sentence Embeddings
• A novel repetition normalized adversarial reward for headline generation
• Dialogue Inspectional Summarization with Factual Inconsistency Awareness
• SgSum: Transforming Multi-document Summarization into Sub-graph Selection
• Improving Graph-based Sentence Ordering with Iteratively Predicted Pairwise Orderings
• TLDR9+: A Large Scale Resource for Extreme Summarization of Social Media Posts
• Controllable Neural Dialogue Summarization with Personal Named Entity Planning
• MiRANews: Dataset and Benchmarks for Multi-Resource-Assisted News Summarization
• From web crawled text to project descriptions: automatic summarizing of social innovation projects
• Topic-Aware Contrastive Learning for Abstractive Dialogue Summarization
• Medically Aware GPT-3 as a Data Generator for Medical Dialogue Summarization
• Multiplex Graph Neural Network for Extractive Text Summarization
• Enhanced Seq2Seq Autoencoder via Contrastive Learning for Abstractive Text Summarization
• Keeping Notes: Conditional Natural Language Generation with a Scratchpad Mechanism
• DESCGEN: A Distantly Supervised Dataset for Generating Abstractive Entity Descriptions