Knowledge question and answer processing method and device based on large model

By collaborating with a large model and a vector knowledge base, user questions are broken down into sub-questions and answers are retrieved from the vector knowledge base. This solves the shortcomings of existing technologies in decomposing complex questions and semantic retrieval, and generates logically clear and professional knowledge-based question-and-answer results.

CN120822611APending Publication Date: 2025-10-21BEIJING UNISOUND INFORMATION TECH CO LTD +7
0 Cites 0 Cited by

Patent Information

Application Number
CN202510960982.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-07-12
Publication Date
2025-10-21

AI Technical Summary

Technical Problem

Existing knowledge-based question answering technologies have shortcomings in complex question decomposition, semantic knowledge retrieval and integration, and the collaborative application of large models and professional knowledge bases, and cannot meet the requirements of accuracy, comprehensiveness and professionalism in question answering.

Method used

By employing a knowledge-based question-answering method based on a large model, user questions are assembled with prompt templates for decomposing questions to generate sub-questions. Relevant content is then retrieved from a vector knowledge base, and answers are generated using a large model. This process is iterated until a complete answer is obtained. The accuracy is verified by combining the TF-IDF algorithm and the BERT model.

Benefits of technology

It enables flexible breakdown of complex problems and accurate knowledge retrieval, generating coherent and professionally compliant answers, thereby improving the logical integrity of the question-and-answer system and user satisfaction.

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

Abstract

The invention discloses a knowledge question and answer processing method and device based on a large model, and relates to the field of natural language processing. The method comprises the following steps: assembling a user question and disassembling a prompt big model to obtain sub-questions; retrieving related contents of the sub-questions, and then assembling an enlargement model to generate questions and answers; circulating until all sub-answers are obtained; the original questions, the sub-questions, the answers and the summary prompt are assembled, and summary content is generated and returned. The device comprises a question splitting module, a content retrieval module and the like. Through cooperation of the large model and the vector knowledge base, complex question disassembly, accurate knowledge retrieval and answer integration are achieved, the accuracy and comprehensiveness of knowledge questions and answers are improved, and the method is suitable for scenes such as intelligent customer service and online education.
Need to check novelty before this filing date? Find Prior Art