Pre-training for Abstractive Document Summarization by Reinstating Source Text
arXiv:2004.01853
Abstract
Abstractive document summarization is usually modeled as a sequence-to-sequence (Seq2Seq) learning problem. Unfortunately, training large Seq2Seq based summarization models on limited supervised summarization data is challenging. This paper presents three pre-training objectives which allow us to pre-train a Seq2Seq based abstractive summarization model on unlabeled text. The main idea is that, given an input text artificially constructed from a document, a model is pre-trained to reinstate the original document. These objectives include sentence reordering, next sentence generation, and masked document generation, which have close relations with the abstractive document summarization task. Experiments on two benchmark summarization datasets (i.e., CNN/DailyMail and New York Times) show that all three objectives can improve performance upon baselines. Compared to models pre-trained on large-scale data (more than 160GB), our method, with only 19GB text for pre-training, achieves comparable results, which demonstrates its effectiveness.
arXiv commentEMNLP2020 Camera-Ready, 15 pages
References in corpus (9)
- Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer
- Teaching Machines to Read and Comprehend
- A Deep Reinforced Model for Abstractive Summarization
- PEGASUS: Pre-training with Extracted Gap-sentences for Abstractive Summarization
- SQuAD: 100,000+ Questions for Machine Comprehension of Text
- SummaRuNNer: A Recurrent Neural Network based Sequence Model for Extractive Summarization of Documents
- MASS: Masked Sequence to Sequence Pre-training for Language Generation
- A large annotated corpus for learning natural language inference
- ProphetNet: Predicting Future N-gram for Sequence-to-Sequence Pre-training