Multi-context semantic recognition and understanding method based on large language model

By employing a multi-context semantic recognition and understanding method based on a large language model, which integrates semantic unit sequences and contextual nesting vectors, and combines prefix modulation mechanisms and neural networks, the problem of insufficient context modeling in multilingual semantic recognition is solved. This achieves high-precision intent recognition and multilingual adaptation, and enhances the system's intelligent response capabilities.

CN121234940APending Publication Date: 2025-12-30CHENGDU YUNDA ZHIYE TECH CO LTD
0 Cites 3 Cited by

Patent Information

Application Number
CN202511322910.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-09-16
Publication Date
2025-12-30

AI Technical Summary

Technical Problem

Existing multilingual semantic recognition methods exhibit insufficient generalization and semantic drift when dealing with complex interaction modes such as multi-turn dialogues, cross-lingual mixed expressions, and dynamic context switching. They are difficult to establish an effective context modeling mechanism, resulting in unstable intent recognition and low semantic accuracy, and lack of the ability to flexibly adapt to changes in context.

Method used

We adopt a multi-context semantic recognition and understanding method based on a large language model, which integrates semantic unit sequences, contextual nesting vectors, language labels and historical intent states. We guide semantic modeling through a prefix regulation mechanism, construct contextual prior representations, and use gated recurrent neural networks and feedforward neural networks for intent recognition and slot filling to achieve high robustness and high accuracy semantic understanding in multilingual scenarios.

Benefits of technology

It enhances semantic modeling capabilities in multilingual environments, ensures the accuracy of intent recognition and multilingual adaptability, realizes closed-loop processing from semantic recognition to service response, and enhances the system's intelligent response capabilities and user interaction experience.

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

Abstract

The invention discloses a multi-context semantic recognition and understanding method based on a large language model. The method comprises the following steps: S1, generating a semantic unit sequence; s2, constructing a context nested vector sequence; s3, constructing context priori representation, and generating a semantic representation sequence; s4, outputting a state vector of the semantic path by adopting a gating loop unit, obtaining a matching degree score according to a feedforward neural network, and determining a deliberate map tag and an alternative intention tag; s5, slot field extraction and semantic filling are completed, and a structured semantic task unit is generated; and S6, completing semantic recognition and service response closed loop. According to the method, by introducing a prefix regulation and control mechanism and a multi-context semantic modeling structure, the accuracy of intention recognition in multiple rounds of dialogues and the consistency of the context generated in response are remarkably improved, and the method is suitable for natural language understanding scenes of multi-language intelligent customer service, cross-context man-machine interaction and complex task driving.
Need to check novelty before this filing date? Find Prior Art