Google patents search engine that breaks tasks into subtasks
Google has filed a patent for a search system that uses machine learning to split a user's request into smaller, specific subtasks. The system searches for content related to each subtask and displays the results on a page, ranking them based on how likely users are to interact with them.
In the filing’s own words
A task-based search system is described. The computing system can receive, from a user device, a first user query associated with a task. The system can determine, using one or more machine-learned models, a first subtask having a first interaction score and a second subtask having a second user interaction score. Additionally, the system can perform a first query search for the first subtask to obtain a first content item associated with the first subtask. Moreover, the system can perform a second query search for the second subtask to obtain a second content item associated with the second subtask. Furthermore, the system can cause a presentation, on a display of the user device, of the first content item and the second content item. The first content item can be displayed above the second content item based on the first user interaction score being higher than the second user interaction score.
What exactly they claim (claim 1)
1. A computing system, comprising: one or more processors; and one or more non-transitory computer-readable media storing instructions that are executable by the one or more processors to cause the computing system to perform operations, the operations comprising: receiving, from a user device, a first user query; determining, using a multi-task machine-learned model, that the first user query is associated with a first subtask and a second subtask of a task; performing, using the multi-task machine-learned model, a first query search for the first subtask to obtain a first content item associated with the first subtask; performing, using the multi-task machine-learned model, a second query search for the second subtask to obtain a second content item associated with the second subtask; causing a presentation, on a display of the user device, of a search result page, the search result page displaying the first content item and the second content item; modifying a user interaction score associated with the second subtask in response to receiving a user interaction with the second content item; and updating, based on the user interaction score, the search result page to include a third content item as the search result page is being scrolled, the third content item being associated with the second subtask.
Facts
- Company
- Google (Google LLC)
- Official title
- Search System Having Task-Based Machined-Learned Models
- Publication number
- US20260288891A1
- Application number
- 19696064
- Filed
- Jun 2, 2026
- Published
- Sep 24, 2026
- Inventors
- 3
- Claims
- 20
Classified under
G06F 16/954 Physics & computingG06F 16/9538 Physics & computingMore from Google in AI & software
- Google patents video search that answers questions using AI descriptions Sep 15, 2026
- Google patents photo tool that removes people and their gear Sep 15, 2026
- Google patents AI that writes live meeting notes and summaries Sep 15, 2026
- Google patents AI tool that turns messy emergency notes into clear messages Sep 3, 2026
- Google patents AI that clicks through apps to finish tasks for you Sep 24, 2026
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.