According to Tianyancha's property clue information, on May 9, the patent for "Domain Question and Answer Big Model Training and Question and Answer Method, Related Equipment and Program Products" applied by iFLYTEK Co., Ltd. was announced. The abstract shows that this application introduces a referee model that can give preference scores to input answers and evaluate the quality of the answers. Domain question and answer data is extracted based on the domain knowledge base. When the initial big model is iteratively trained, the referee big model scores the first answer corresponding to the domain question generated by the initial big model and the second answer in the domain question and answer data. Based on the scoring results, the answers and domain questions that meet the preference requirements are selected to form the target training data. The initial big model is trained with the target training data to obtain the final domain question and answer big model, which uses high-quality domain question and answer training data for training. This application does not require manual proofreading of the extracted domain question and answer data to obtain high-quality target training data, saving labor costs and improving training efficiency.