Jointly Extracting and Compressing Documents with Summary State Representations
This work addresses the problem of generating concise and informative summaries for users, offering a balance between extractive and abstractive approaches, though it is incremental in nature.
The authors tackled text summarization by developing a neural model that extracts sentences and then compresses them, achieving state-of-the-art results on CNN/DailyMail and Newsroom datasets with improved performance over existing methods.
We present a new neural model for text summarization that first extracts sentences from a document and then compresses them. The proposed model offers a balance that sidesteps the difficulties in abstractive methods while generating more concise summaries than extractive methods. In addition, our model dynamically determines the length of the output summary based on the gold summaries it observes during training and does not require length constraints typical to extractive summarization. The model achieves state-of-the-art results on the CNN/DailyMail and Newsroom datasets, improving over current extractive and abstractive methods. Human evaluations demonstrate that our model generates concise and informative summaries. We also make available a new dataset of oracle compressive summaries derived automatically from the CNN/DailyMail reference summaries.