Iterative Document Representation Learning Towards Summarization with Polishing
This work addresses the need for better summarization quality in NLP applications, offering an incremental improvement over existing methods.
The paper tackles the problem of sub-optimal document representations in extractive text summarization by proposing an iterative model that polishes representations over multiple passes, significantly outperforming state-of-the-art systems on CNN/DailyMail and DUC2002 datasets.
In this paper, we introduce Iterative Text Summarization (ITS), an iteration-based model for supervised extractive text summarization, inspired by the observation that it is often necessary for a human to read an article multiple times in order to fully understand and summarize its contents. Current summarization approaches read through a document only once to generate a document representation, resulting in a sub-optimal representation. To address this issue we introduce a model which iteratively polishes the document representation on many passes through the document. As part of our model, we also introduce a selective reading mechanism that decides more accurately the extent to which each sentence in the model should be updated. Experimental results on the CNN/DailyMail and DUC2002 datasets demonstrate that our model significantly outperforms state-of-the-art extractive systems when evaluated by machines and by humans.