Abstract: Short-term climate prediction of precipitation in the basin is imperative for disaster prevention and reduction in the basin. To further improve the prediction capacity of the new-generation atmospheric general circulation model, IAP AGCM 4.1 by the Institute of Atmospheric Physics of the Chinese Academy of Sciences, on the summer precipitation in the Huaihe and Yangtze River Basins, a regional Empirical Orthogonal Function (EOF) correction scheme suitable for the two river basins is established based on the analysis of the regional characteristics of the summer precipitation using the rotating EOF (REOF) method. Furthermore, the new scheme was validated based on 30-year (1981–2010) hindcasts of the IAP AGCM 4.1 climate model in the Huaihe and Yangtze River Basins. Results reveal that the new correction method significantly improves the prediction of summer precipitation in the Huaihe River Basin, with the average correlation coefficient of the Huaihe River Basin increasing from 0.03 to 0.22. Moreover, the seasonal precipitation forecast of the Yangtze River Basin is also significantly improved, with the average correlation coefficient increasing from −0.05 to 0.24. The results of the new correction method are notably better than those of previous methods based on whole-basin data, confirming that using strong local precipitation characteristics to determine the correction area based on REOF analysis can improve the effect and stability of EOF correction. The proposed method can potentially be applied to other basins as well.