Selective Encoding for Abstractive Sentence Summarization
This work addresses the problem of generating concise summaries for sentences, which is incremental as it builds on existing sequence-to-sequence methods.
The authors tackled abstractive sentence summarization by proposing a selective encoding model that extends the sequence-to-sequence framework, resulting in improved performance over state-of-the-art baselines on datasets like English Gigaword, DUC 2004, and MSR.
We propose a selective encoding model to extend the sequence-to-sequence framework for abstractive sentence summarization. It consists of a sentence encoder, a selective gate network, and an attention equipped decoder. The sentence encoder and decoder are built with recurrent neural networks. The selective gate network constructs a second level sentence representation by controlling the information flow from encoder to decoder. The second level representation is tailored for sentence summarization task, which leads to better performance. We evaluate our model on the English Gigaword, DUC 2004 and MSR abstractive sentence summarization datasets. The experimental results show that the proposed selective encoding model outperforms the state-of-the-art baseline models.