Asphalt pavement construction requires accurate, rapid, and continuous quality evaluation. As a primary defect appearing in construction, segregation significantly reduces the pavement's mechanical properties and service life, making it become a hot spot in asphalt pavement construction research. This review aims to provide a state-of-the-art review about evaluating and quantifying hot mix asphalt segregation and identify the opportunities and challenges in the quality evaluation of automatic asphalt pavement construction. This review contained text mining findings and critical surveys. Text mining was used to analyze the co-occurrence relationship of research keywords and the change of research heats. Based on the text mining findings, we critically reviewed the data collection, data processing, and quantification indexes of the subjective, direct-objective, and indirect-objective segregation evaluation methods. The remaining deficiencies and possible research directions have been discussed.