<p>Enhancing transportation sector carbon emissions efficiency (TSCEE) is crucial for achieving low-carbon transformation and sustainable development. Using a global super-efficiency EBM model with undesirable outputs, we empirically measure TSCEE and employ methods such as the Dagum Gini coefficient, exploratory spatial data analysis, kernel density estimation, and traditional/spatial Markov chains to investigate its regional disparities and dynamic evolution characteristics. Furthermore, the optimal parameter-based geographic detector (OPGD) model is applied to identify the critical factors driving the spatial differentiation of TSCEE from a heterogeneity perspective. The findings reveal four key insights. First, China’s overall TSCEE remains at a relatively low level, primarily due to high redundancy rates of energy input and carbon emissions. Second, TSCEE exhibits significant spatial inequality, with inter-regional differences being the major contributor to overall disparities. Spatial autocorrelation analysis shows a gradually strengthening positive spatial agglomeration pattern. Third, the kernel density curves of TSCEE exhibit a right-trailing and polarization phenomenon. Provincial TSCEE tends to maintain its original state, with limited potential for leapfrogging improvements. Fourth, technological progress is the most significant determinant of TSCEE, and the heterogeneity analysis reveals significant regional variations in the driving forces behind TSCEE. The interactions between various factors, particularly the interactions between technological progress and other factors, significantly enhance the explanatory powers of the spatial differentiation of TSCEE. This study provides a comprehensive analysis of TSCEE in China, offering valuable insights for policymakers to design region-specific strategies and promote low-carbon transitions in the transportation sector.</p>