Abstract: Landslide susceptibility assessment is a crucial component of landslide risk management, effectively guiding disaster prevention and mitigation efforts. However, the accuracy of landslide susceptibility assessments is constrained by various factors, and current research on optimizing negative sample sampling strategies based on slope units remains relatively limited. This study, focuses on Bazhong City as the research area, incorporates eleven conditioning factors including elevation, relief, and annual average rainfall to develop a geographically weighted regression - random forest (GWR-RF) coupling model. This model optimize the negative sampling strategy by comparing it against traditional random sampling across the entire area. The results indicate the following: (1) Random sampling from the entire area leads to significant disparities in susceptibility assessments, accompanied by a relatively diminished accuracy, rendering it unsuitable for slope unit-based assessments. (2) The coupled GWR-RF model demonstrates spatial variations in landslide susceptibility, predominantly distributing in the Enyang, Bazhou, Pingchang Counties, and the central - southern region of Nanjiang County. The proposed GWR-RF coupled model improves the accuracy of landslide susceptibility assessments by optimizing the negative sample sampling strategy, providing a scientific basis for landslide disaster prevention and mitigation in the Bazhong region.