Text generation method and device, electronic equipment and storage medium

By splitting complex query requests into subquery requests, and introducing prompt word generation and text generation modules dedicated to specific scenarios in RAG technology, the problem of low iteration update efficiency in existing RAG technology in different application scenarios is solved, and more efficient text generation device update is achieved.

CN119719356AActive Publication Date: 2025-03-28UC MOBILE CHINA CO LTD
6 Cites 1 Cited by

Patent Information

Application Number
CN202510229660.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-02-28
Publication Date
2025-03-28
Estimated Expiration
2045-02-28

AI Technical Summary

Technical Problem

The existing search enhanced generation (RAG) technology needs to be deployed and iteratively updated in different application scenarios, resulting in low efficiency of iterative updates.

Method used

The complex query request is split into multiple subquery requests through the split module. The search module retrieves corresponding reference information from the knowledge base. The prompt word generation module generates prompt words based on the reference information, and generates corresponding text through the text generation module. This scheme uses the split module and the search module for multiple text generation scenarios, while the prompt word generation module and the text generation module are dedicated to specific scenarios.

Benefits of technology

It improves the iterative update efficiency of the text generation device in different application scenarios, and reduces the need for duplicate updates for split modules and search modules.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The embodiment of the invention provides a text generation method and device, electronic equipment and a storage medium, and the text generation device comprises a splitting module which is used for splitting a first query request from user equipment into a plurality of sub-query requests; the retrieval module is used for retrieving reference information corresponding to the first query request and the sub-query request from a knowledge base; the cue word generation module is used for generating a first cue word according to the reference information; the text generation module is used for inputting the first prompt word into a text generation model and generating a text corresponding to the first query request through the text generation model; the prompt word generation module and the text generation module are specially used for a text generation scene corresponding to the first query request, and the splitting module and the retrieval module are used for a plurality of text generation scenes. According to the scheme, the iterative updating efficiency of the text generation device can be improved.
Need to check novelty before this filing date? Find Prior Art