/Patent Tracker
GooglePatent applicationAI & softwareInterest 6/10

Google patents AI method to split complex questions into smaller searches

This filing describes a system that uses a large language model to break down a complex user question into several smaller, specific search queries. The system filters these subqueries for relevance and variety, retrieves search results for each, and combines them to create a single, comprehensive answer.

Why it mattersIt offers a technical approach to making AI assistants better at handling multi-part questions by ensuring they search for all necessary pieces of information before answering.

In the filing’s own words

Implementations leverage a generative model (e.g., a large language model (LLM)) to generate a plurality of candidate subqueries for multifaceted natural language (NL) based input, where each of the candidate subqueries is potentially directed to a facet or problem of the multifaceted NL based input. Those implementations further select, from the plurality of candidate subqueries and using one or more evaluation metrics, a subset of the candidate queries. Those implementations further, in response to selecting the subset of the candidate queries, obtain, for each of the candidate subqueries of the selected subset, at least one corresponding search result. Those implementations further generate a response to the NL based input based on the corresponding search results for the candidate subqueries of the subset, and cause the response to be rendered responsive to the NL based input.

What exactly they claim (claim 1)

1. A method implemented by one or more processors, the method comprising: receiving natural language (NL) based input associated with a client device; in response to receiving the NL based input: generating a plurality of candidate subqueries for the NL based input based on processing, using a generative model, a subquery generation prompt that is based on the NL based input; selecting, from the plurality of candidate subqueries and using one or more evaluation metrics, a subset of the candidate subqueries generated using the generative model, wherein the one or more evaluation metrics that are utilized in selecting the subset of the candidate subqueries evaluate diversity among the candidate subqueries and/or evaluate relevance of each of the candidate subqueries to the NL based input; obtaining, for each of the candidate subqueries of the subset, at least one corresponding search result; and processing the corresponding search results for the candidate subqueries of the subset, using the generative model or an additional generative model, to generate a response to the NL based input; and causing the response to be rendered at the client device responsive to the NL based input.

Facts

Company
Google (GOOGLE LLC)
Official title
UTILIZING LARGE LANGUAGE MODEL (LLM) IN RESPONDING TO MULTIFACETED QUERIES
Publication number
US20260277895A1
Application number
19684688
Filed
May 21, 2026
Published
Sep 17, 2026
Inventors
3
Claims
18

Classified under

G06F 16/24535 Physics & computingG06F 16/24542 Physics & computingG06F 40/40 Physics & computing
Read the full filing (official PDF) ↗ Open in the Patent Tracker

More from Google in AI & software

Summaries are simplified from the filing’s own abstract and first claim. A patent filing is not a product announcement.
Patent data and drawings: U.S. Patent and Trademark Office (USPTO). Nebula is not affiliated with or endorsed by the USPTO.