Don't Give Me the Details, Just the Summary! Topic-Aware Convolutional Neural Networks for Extreme Summarization

Abstract

We introduce extreme summarization, a new single-document summarization task which does not favor extractive strategies and calls for an abstractive modeling approach. The idea is to create a short, one-sentence news summary answering the question "What is the article about?". We collect a real-world, large-scale dataset for this task by harvesting online articles from the British Broadcasting Corporation (BBC). We propose a novel abstractive model which is conditioned on the article's topics and based entirely on convolutional neural networks. We demonstrate experimentally that this architecture captures long-range dependencies in a document and recognizes pertinent content, outperforming an oracle extractive system and state-of-the-art abstractive approaches when evaluated automatically and by humans.

11, 2018 Conference on Empirical Methods in Natural Language Processing, EMNLP 2018

References in corpus (3)

• Convolutional Sequence to Sequence Learning
• SummaRuNNer: A Recurrent Neural Network based Sequence Model for Extractive Summarization of Documents
• Neural Extractive Summarization with Side Information

Cited by in corpus (16)

• Multitask Prompted Training Enables Zero-Shot Task Generalization
• fairseq: A Fast, Extensible Toolkit for Sequence Modeling
• Pre-trained Summarization Distillation
• Sticking to the Facts: Confident Decoding for Faithful Data-to-Text Generation
• Efficient Adaptation of Pretrained Transformers for Abstractive Summarization
• ABNIRML: Analyzing the Behavior of Neural IR Models
• BERT Fine-tuning For Arabic Text Summarization
• Liputan6: A Large-scale Indonesian Dataset for Text Summarization
• Improving Non-autoregressive Generation with Mixup Training
• BEAMetrics: A Benchmark for Language Generation Evaluation Evaluation
• MRCBert: A Machine Reading ComprehensionApproach for Unsupervised Summarization
• Reference and Document Aware Semantic Evaluation Methods for Korean Language Summarization
• Spinning Sequence-to-Sequence Models with Meta-Backdoors
• Noised Consistency Training for Text Summarization
• Resurrecting Submodularity for Neural Text Generation
• Open4Business(O4B): An Open Access Dataset for Summarizing Business Documents