A large language model knowledge question answering method based on semantic parsing correction

By constructing a semantic parsing and correction method based on knowledge graphs, generating fine-tuning sample sets and designing structured instruction templates, the problem of knowledge statics and path generation fragmentation in multi-hop question-answering scenarios of large language models is solved, improving the accuracy and interpretability of answers and realizing the deep integration of knowledge and language models.

CN120764680BActive Publication Date: 2026-07-24HUAZHONG UNIV OF SCI & TECH
2 Cites 0 Cited by

Patent Information

Application Number
CN202510871832.9
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2025-06-26
Publication Date
2026-07-24
Estimated Expiration
2045-06-26

AI Technical Summary

Technical Problem

When dealing with multi-hop question-answering scenarios, large language models face limitations of knowledge staticity and timeliness, lack effective knowledge graph interaction interfaces, and are disconnected from natural language understanding, making it difficult to achieve credibility and controllability. Furthermore, the generated answers are prone to illusions and logical errors.

Method used

By constructing a semantic parsing correction method based on knowledge graphs, a fine-tuning sample set is generated and a structured instruction template is designed to guide the execution path generation task of the large language model. Combined with supervised training and knowledge graph query to verify the answer, illusory paths are eliminated and highly credible answers are generated.

Benefits of technology

It improves the accuracy and robustness of large language models in multi-hop reasoning tasks, enhances the interpretability and controllability of answers, achieves deep integration of structured knowledge and language models, and improves reasoning ability and answer accuracy in complex question-answering tasks.

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

Abstract

The application discloses a large language model knowledge question and answer method based on semantic analysis correction, and belongs to the natural language reasoning field. The method extracts key entities from a natural language question and matches target entities in a knowledge graph, and mines multiple knowledge paths in the knowledge graph for model training. A prompt instruction and a knowledge path are used to supervise the training of a large language model, so that the large language model has the ability to generate a multi-hop reasoning path that conforms to the structure of the knowledge graph. Based on the trained model, knowledge path decoding is performed, the natural language question is converted into multiple candidate reasoning paths, and is converted into a logical query statement to be executed in the knowledge graph, thereby completing the fact verification and completion of the path. The high-quality path verified by the knowledge graph and the natural language question are input into the large language model to obtain an accurate and interpretable answer. The method realizes the deep fusion of structured knowledge and the language model, and improves the reasoning ability, answer accuracy and interpretability in complex question and answer tasks.
Need to check novelty before this filing date? Find Prior Art