Systems and methods for generating query responses

CA3281836A1Pending Publication Date: 2024-09-26SHOPIFY INC
0 Cites 0 Cited by

Patent Information

Application Number
CA3281836
Authority / Receiving Office
CA · CA
Patent Type
Applications
Current Assignee / Owner
Priority Date
2023-05-31
Filing Date
2023-10-06
Publication Date
2024-09-26

AI Technical Summary

Technical Problem

Existing query resolution systems using large language models (LLMs) face challenges such as generating stale responses due to frequent platform changes, inability to distinguish semantically similar queries, and inefficiency in processing large datasets, leading to suboptimal query response generation.

Method used

A system that clusters queries based on semantic similarity using LLMs, allowing for the identification of dissimilar queries and generating responses by matching incoming queries to specific clusters, thereby refining response generation and reducing the need for frequent re-training.

Benefits of technology

This approach enhances the accuracy and relevance of query responses by distinguishing between semantically distinct queries and reduces the resource intensity of re-training, ensuring timely and contextually appropriate answers.

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

Abstract

A computer-implemented method is disclosed. The method includes: obtaining at least one query; clustering a set comprising the at least one query into first clusters; for each first cluster, identifying, by a large language model (LLM), queries in the cluster that are semantically dissimilar; clustering the queries identified as semantically dissimilar into one or more second clusters; receiving an incoming query; matching the incoming query to a particular cluster from the first or second clusters; obtaining one or more generated response messages based on providing, to the LLM, data associated with the particular cluster for the incoming query.
Need to check novelty before this filing date? Find Prior Art