A distributed intelligent question-answering method based on a novel federated knowledge learning model

Abstract

This invention relates to a distributed intelligent question-answering method based on a novel federated knowledge learning model, belonging to the field of natural language processing. The method includes: each client fine-tuning its private downstream task model using local privacy data, while the server fine-tunes the global model using a public dataset to enhance its knowledge understanding capabilities; a central server decomposes the task flow into required requirements and distributes these requirements to the clients; each client generates corresponding knowledge fragments based on the received requirements and uploads them to the server; the server aggregates the fragmented knowledge received from the clients to form complete task-aiding knowledge; and the server's large model combines the task-aiding knowledge to generate a task response. This invention introduces task decomposition, knowledge aggregation, and de-identification modules, ensuring not only high performance but also robust privacy protection.

Images