Invention Grant
- Patent Title: Generating query variants using a trained generative model
-
Application No.: US16609318Application Date: 2018-04-27
-
Publication No.: US11663201B2Publication Date: 2023-05-30
- Inventor: Jyrki Alakuijala , Christian Buck , Jannis Bulian , Massimiliano Ciaramita , Wojciech Gajewski , Andrea Gesmundo , Neil Houlsby , Wei Wang
- Applicant: Google LLC
- Applicant Address: US CA Mountain View
- Assignee: GOOGLE LLC
- Current Assignee: GOOGLE LLC
- Current Assignee Address: US CA Mountain View
- Agency: Gray Ice Higdon
- International Application: PCT/US2018/029834 2018.04.27
- International Announcement: WO2018/200979A 2018.11.01
- Date entered country: 2019-10-29
- Main IPC: G06F16/242
- IPC: G06F16/242 ; G06N3/08 ; G06N3/047 ; G06N3/044

Abstract:
Systems, methods, and computer readable media related to generating query variants for a submitted query. In many implementations, the query variants are generated utilizing a generative model. A generative model is productive, in that it can be utilized to actively generate a variant of a query based on application of tokens of the query to the generative model, and optionally based on application of additional input features to the generative model.
Information query