Abstract: Soil salinization can inhibit crop growth, diminish soil fertility, and cause ecological degradation, thereby leading to severe economic losses. The remote sensing technology provides real-time data with high spatiotemporal resolution for identifying, grading, and dynamically monitoring field-scale soil salinization. Focusing on current hot research topics of field-scale soil salinization, including multi-source and multi-scale integration, time series analysis, and machine learning/deep learning inversion and interpretability, this study reviewed the application of the remote sensing monitoring technology in field-scale soil salinization research, followed by a comprehensive analysis of the current status of agricultural soil salinization in fields. Based on domestic and international research advances, this study summarized the mechanisms and modeling for salinization monitoring, the methods for dynamic field monitoring, the application of field-scale soil salinization research methodologies, as well as the advantages and limitations of various types of sensors. However, challenges persist in the field-scale soil salinization research, including the prominent presence of mixed pixels, strong salinity heterogeneity induced by microtopography and irrigation, as well as the disruption by agricultural activities on water-salt migration. Therefore, from the perspective of artificial intelligence, this study suggests optimizing the scale transform and regional adaptation, developing physical models, and integrating multi-source data to enhance model robustness across different land use types and ecological zones. Overall, this study provides insights for future field-scale salinization monitoring and treatment.