Apparatus and method for processing natural language using boosting keyword sets and suppressing keyword sets

The method enhances LLM models by using boosting and suppressing keyword sets to control sentence generation, improving processing speed and keyword relevance in natural language processing.

JP7917210B2Active Publication Date: 2026-09-08SIONIC AI INC
3 Cites 0 Cited by

Patent Information

Application Number
JP2025066746
Authority / Receiving Office
JP · JP
Patent Type
Patents
Current Assignee / Owner
Priority Date
2024-06-12
Filing Date
2025-04-15
Publication Date
2026-09-08
Estimated Expiration
2045-04-15

AI Technical Summary

Technical Problem

Conventional Large Language Model (LLM) models trained in foreign languages face issues with foreign language generation and undesired keyword generation due to user's inability to modify the training dataset, necessitating a technique to control sentence generation based on desired keywords.

Method used

A natural language processing method using boosting and suppressing keyword sets, where a boosting keyword set enhances generation and a suppression keyword set suppresses generation, implemented through an artificial neural network model to adjust generation probabilities based on keyword sets.

Benefits of technology

Improves natural language processing speed, adjusts generation probabilities, increases the likelihood of desired keywords, and reduces undesired keywords in generated text.

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

Abstract

To provide an apparatus and method of processing a natural language using a boosting keyword set and suppressing keyword set.SOLUTION: A natural language processing method performed in an electronic device including at least one processor and at least one memory storing commands to be executed by the at least one processor includes the steps of: acquiring a boosting keyword set including at least one boosting keyword that is an object of generation boost when generating a sentence using an artificial neural network model; acquiring a suppressing keyword set including at least one suppressing keyword that is an object of generation suppression when generating a sentence using the artificial neural network model; and generating sentences through the artificial neural network model based on the boosting keyword set and the suppressing keyword set.SELECTED DRAWING: Figure 4
Need to check novelty before this filing date? Find Prior Art