<p>This study investigates the spatial characteristics, evolutionary patterns, and influencing factors of carbon emission spillover effects, aiming to advance China’s carbon reduction initiatives and foster regional coordinated development. Based on panel data from 282 Chinese cities (2007–2022), we first measure urban carbon emission efficiency using the Data Envelopment Analysis model and construct a spatial correlation network of carbon emission efficiency through a gravity model. Subsequently, we analyze the spatial features and evolutionary regularities of this network from macro and micro perspectives using network topology metrics and motif structure analysis. Finally, Temporal Exponential Random Graph Models are employed to identify factors influencing the carbon emission efficiency correlation network. Key findings include: ① The spatial correlation network exhibits a multilayered continuum from strong to weak ties with distinctive “key node effects”. ②At the micro level, the correlation network has the basic structure of “multi-center” and “gradient transfer”, and it can be seen that the network structure has room for further development. ③ The TERGM results show the influence of various endogenous structural variables on the development of the carbon emission efficiency correlation network. ④ Different economic factors have significant differences in the spillover effect of carbon emission efficiency, and after distinguishing the functions and policies of cities, the impact of innovation level and green development level on the correlation of carbon emission efficiency has changed significantly. Accordingly, this study provides valuable suggestions for policy makers, and offers new research perspectives and practical guidance for carbon emission reduction work through aspects such as urban coordinated development and governance, optimization paths of carbon emission efficiency, and optimized development of spillover effects.</p>