Bottom-Up Abstractive Summarization

Neural network-based methods for abstractive summarization produce outputs that are more fluent than other techniques, but which can be poor at content selection. This work proposes a simple technique for addressing this issue: use a data-efficient content selector to over-determine phrases in a source document that should be part of the summary. We use this selector as a bottom-up attention step to constrain the model to likely phrases. We show that this approach improves the ability to compress text, while still generating fluent summaries. This two-step process is both simpler and higher performing than other end-to-end content selection models, leading to significant improvements on ROUGE for both the CNN-DM and NYT corpus. Furthermore, the content selector can be trained with as little as 1,000 sentences, making it easy to transfer a trained summarizer to a new domain.

PDF Abstract EMNLP 2018 PDF EMNLP 2018 Abstract
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Document Summarization CNN / Daily Mail Bottom-Up Sum PPL 32.75 # 2
ROUGE-1 41.22 # 12
ROUGE-2 18.68 # 13
ROUGE-L 38.34 # 12
Abstractive Text Summarization CNN / Daily Mail Bottom-Up Summarization ROUGE-1 41.22 # 25
ROUGE-2 18.68 # 24
ROUGE-L 38.34 # 23
Multi-Document Summarization Multi-News CopyTransformer ROUGE-2 14.03 # 6
ROUGE-1 43.57 # 4
ROUGE-SU4 17.37 # 4
Multi-Document Summarization Multi-News PG-BRNN ROUGE-2 14.19 # 5
ROUGE-1 42.80 # 6
ROUGE-SU4 16.75 # 5

Methods


No methods listed for this paper. Add relevant methods here