A Deep Reinforced Model for Abstractive SummarizationDownload PDF

15 Feb 2018 (modified: 21 Apr 2024)ICLR 2018 Conference Blind SubmissionReaders: Everyone
Abstract: Attentional, RNN-based encoder-decoder models for abstractive summarization have achieved good performance on short input and output sequences. For longer documents and summaries however these models often include repetitive and incoherent phrases. We introduce a neural network model with a novel intra-attention that attends over the input and continuously generated output separately, and a new training method that combines standard supervised word prediction and reinforcement learning (RL). Models trained only with supervised learning often exhibit "exposure bias" - they assume ground truth is provided at each step during training. However, when standard word prediction is combined with the global sequence prediction training of RL the resulting summaries become more readable. We evaluate this model on the CNN/Daily Mail and New York Times datasets. Our model obtains a 41.16 ROUGE-1 score on the CNN/Daily Mail dataset, an improvement over previous state-of-the-art models. Human evaluation also shows that our model produces higher quality summaries.
TL;DR: A summarization model combining a new intra-attention and reinforcement learning method to increase summary ROUGE scores and quality for long sequences.
Keywords: deep learning, natural language processing, reinforcement learning, text summarization, sequence generation
Code: [![Papers with Code](/images/pwc_icon.svg) 10 community implementations](https://paperswithcode.com/paper/?openreview=HkAClQgA-)
Data: [CNN/Daily Mail](https://paperswithcode.com/dataset/cnn-daily-mail-1), [New York Times Annotated Corpus](https://paperswithcode.com/dataset/new-york-times-annotated-corpus)
Community Implementations: [![CatalyzeX](/images/catalyzex_icon.svg) 7 code implementations](https://www.catalyzex.com/paper/arxiv:1705.04304/code)
7 Replies