Exploring the multifunctionality of arable land and identifying the spatiotemporal characteristics of its various functions are essential for promoting the coordinated and sustainable use of farmland in Xinjiang. Based on socio-economic data from 14 prefectures in Xinjiang from 2009 to 2020, this study employs a coupling coordination degree model, spatial autocorrelation analysis, and the geographical detector method to measure the coupling coordination degree of the “production-living-ecological” functions of arable land and to analyze its spatiotemporal evolution and driving factors. The results indicate that: (1) From 2009 to 2020, the production, living, and ecological functions of arable land in Xinjiang exhibited varying degrees of spatial and temporal differentiation. The production and living functions showed an upward trend, while the ecological function demonstrated a downward trend. (2) During the study period, the coupling coordination degree of the three functions increased slightly, though no prefecture reached a state of high coordination. Most prefectures remained in a state of imbalance. Spatially, the coordination degree was generally higher in northern Xinjiang than in the southern region, shifting from a monocentric to a polycentric agglomeration pattern. (3) The primary driving factors of the coupling coordination degree were regional policy implementation level, agricultural development level, and industrial development level. The interactions among these factors manifested as both bivariate enhancement and nonlinear enhancement, with a notable transition trend from bivariate to nonlinear enhancement over time.