Generative Adversarial Network for Abstractive Text Summarization
This work addresses text summarization for natural language processing applications, but it is incremental as it adapts existing GAN and reinforcement learning techniques to this task.
The authors tackled abstractive text summarization by proposing a generative adversarial network (GAN) with a reinforcement learning-based generator and a discriminator, achieving competitive ROUGE scores on the CNN/Daily Mail dataset and generating more abstractive, readable, and diverse summaries.
In this paper, we propose an adversarial process for abstractive text summarization, in which we simultaneously train a generative model G and a discriminative model D. In particular, we build the generator G as an agent of reinforcement learning, which takes the raw text as input and predicts the abstractive summarization. We also build a discriminator which attempts to distinguish the generated summary from the ground truth summary. Extensive experiments demonstrate that our model achieves competitive ROUGE scores with the state-of-the-art methods on CNN/Daily Mail dataset. Qualitatively, we show that our model is able to generate more abstractive, readable and diverse summaries.