Systems and methods for performing vector search
AssigneeShopify Inc.
InventorsDaniel BEAUCHAMP, Andreas WINSCHU
Life of this patent
Estimating the term…
Abstract
A computer-implemented method is disclosed. The method includes: receiving a search query for a search of items in a database; providing, to a large language model (LLM), an input including the search query and instructions to generate query enhancement text for the search query; combining the search query and the query enhancement text output by the LLM; obtaining a query vector embedding based on the combined search query and the query enhancement text; and performing a search of an embedding space associated with the database using the query vector embedding to identify one or more search results.
Claims20
A computer-implemented method, comprising:
- receiving a search query for a search of items in a database;
- providing, to a large language model (LLM), an input including the search query and instructions to generate query enhancement text for the search query;
- combining the search query and the query enhancement text output by the LLM;
- obtaining a query vector embedding based on the combined search query and the query enhancement text;
- and
- performing a search of an embedding space associated with the database using the query vector embedding to identify one or more search results.
A computing system, comprising:
- a processor;
- and
- a memory coupled to the processor, the memory storing computer-executable instructions that, when executed by the processor, configure the processor to:
- receive a search query for a search of items in a database;
- provide, to a large language model (LLM), an input including the search query and instructions to generate query enhancement text for the search query;
- combine the search query and the query enhancement text output by the LLM;
- obtain a query vector embedding based on the combined search query and the query enhancement text;
- and
- perform a search of an embedding space associated with the database using the query vector embedding to identify one or more search results.
A non-transitory processor-readable medium storing processor-executable instructions that, when executed by a processor, are to cause the processor to:
- receive a search query for a search of items in a database;
- provide, to a large language model (LLM), an input including the search query and instructions to generate query enhancement text for the search query;
- combine the search query and the query enhancement text output by the LLM;
- obtain a query vector embedding based on the combined search query and the query enhancement text;
- and
- perform a search of an embedding space associated with the database using the query vector embedding to identify one or more search results.
Dependencies are inferred from the claim text.