Abstract: Regional dry–wet variation is a comprehensive reflection of the surface water supply–demand balance driven by climate, and it directly affects regional hydrological processes, agricultural production stability, and ecosystem security. Northeast China, as an important commodity grain production base and ecological barrier in China, plays a crucial role in ensuring national food security and promoting regional sustainable development. Therefore, understanding the characteristics and driving mechanisms of dry–wet variations in this region is of great theoretical and practical significance. To improve the applicability of the MOD16A2GF evapotranspiration dataset in Northeast China and to better characterize regional dry–wet variations, this study was based on evapotranspiration observations from two stations and meteorological data from 116 stations. A linear regression method was applied to correct actual evapotranspiration (ET) and potential evapotranspiration (PET), and a leave-one-year-out cross-validation approach was used to evaluate the accuracy and reliability of the corrected results. Based on the corrected ET and PET, the crop water stress index (CWSI) was calculated to characterize surface dry–wet conditions in a quantitative manner. Furthermore, the spatiotemporal variations of evapotranspiration and dry–wet conditions in Northeast China from 2001 to 2024 were systematically analyzed, and the driving factors were further explored by considering multiple influencing variables. The results showed that: (1) The MOD16A2GF data showed good agreement with observed values in terms of variation trends. Compared with the original data, the root mean square error (RMSE) and Bias were reduced, while the Nash–Sutcliffe efficiency (NSE) was improved, indicating that the correction effectively enhanced the overall data accuracy and reliability. (2) From 2001 to 2024, ET in Northeast China showed a significant increasing trend, with a growth rate of 3.11 mm/a. PET showed a significant decreasing trend, with a decline rate of 3.23 mm/a. Meanwhile, CWSI showed a significant decreasing trend, with an annual decrease of 0.01, indicating an overall wetting trend across the region and a gradual alleviation of water stress conditions. (3) In terms of spatial patterns, ET mainly ranged from 100 to 800 mm and decreased from east to west. PET mainly ranged from 500 to 1,200 mm and decreased from southwest to northeast, while CWSI ranged from 0.1 to 0.9 and exhibited a similar spatial pattern. The highest CWSI values, indicating the most severe water stress, were observed in the southeastern part of the Eastern Four Leagues in Inner Mongolia. From the perspective of spatial trends, most regions showed increasing ET and decreasing PET, indicating a clear and widespread wetting trend across the study area. Moreover, areas with a significant decrease in CWSI largely overlapped with regions where ET significantly increased and PET significantly decreased, further confirming the spatial consistency and robustness of the wetting trend. (4) Precipitation was positively correlated with ET and negatively correlated with PET and CWSI. The increase in precipitation significantly promoted actual evapotranspiration and suppressed potential evapotranspiration demand, highlighting its dominant role in regulating dry–wet variations in Northeast China. In addition, agricultural irrigation increased surface water supply, while vegetation growth regulated the water exchange process between the land surface and the atmosphere. Both factors enhanced actual evapotranspiration and contributed to alleviating regional water stress to a certain extent. The results provide a scientific basis for optimizing farmland water allocation, preventing drought risks, and improving regional water resource management. They are also of great significance for enhancing regional water regulation capacity and supporting sustainable agricultural development under changing climatic conditions.