Automatic writing assessment leverages computer software to provide automated scoring and evaluation services for assessing learner performance, addressing the limitations of traditional human evaluation methods. Manual assessment typically requires expert raters, substantial time, and effort, and is often prone to biases and inconsistences. Automatic assessment aims to overcome these challanges by utilizing advanced machine learning techniques to analyze and assign meaningful scores to learners’ responses. However, designing effective systems that account for diverse parameters while maintaining accuracy and fairness remains a complex task. Over the past years, significant progress has been made in developing state-of-the-art approaches for automatic text grading. This paper presents a systematic literature review of automatic text scoring, offering a comprehensive overview of recent advancement, highlighting the techniques, datasets, and evaluation metrics employed, and analyzing the limitations of current studies.