A Comprehensive Review of Recent Text Summarization Models

Abstract

The increasing volume of textual data necessitates the development of efficient solutions for its utilization. Therefore, automatic text summarization (ATS) has become one of the most important tasks in natural language processing (NLP). Text summarization are classified as extractive, abstract, and hybrid models. While advanced transformer architectures and large language models (LLMs) have led to the development of ATS models that exploit these techniques. The challenges such as semantic coherence, objectivity, and particularity in long or multi-document texts, limit their performance. Many works have studied and reviewed ATS models, but most have focused on traditional models. Therefore, this paper, provides a comprehensive review of the recent text summarization models. It also discusses the evaluation processes for generated summaries, including dimensions, metrics, and evaluation datasets. Finally, it highlights future research directions in the field of text summarization.