The large systematic biases in coupled model impact the seasonal prediction results. With the motivation to reduce the influence of coupled model biases in seasonal prediction, Singular Value Decomposition (SVD) method was applied in our study to improve the ability to predict flood season precipitation. Based on the coupled climate model CAS-ESM-C, we conducted ensemble seasonal prediction experiments from 1982 to 2017, with the initial conditions provided by the assimilation system. The prediction system was integrated from March to August each year with a focus on the June to August precipitation in China. The results showed that the prediction skills for anomalous summer precipitation were very low without bias corrections. However, the system well predicted the interannual variability of large-scale atmospheric circulation systems that are associated with anomalous summer precipitation. We used SVD method to reduce the pattern-dependent precipitation errors by replacing prediction pattern with observation pattern, the predictive skill for precipitation was dramatically improved. Results demonstrate that correction method is a viable tool to reduce systematic biases in coupled model predictions.