Automatic text summarization, the automated process of shortening a text while reserving the main ideas of the document(s), is a critical research area in natural language processing. The aim of this literature review is to survey the recent work on neural-based models in automatic text summarization. We examine in detail ten state-of-the-art neural-based summarizers: five abstractive models and five extractive models. In addition, we discuss the related techniques that can be applied to the summarization tasks and present promising paths for future research in neural-based summarization.