Changeflow GovPing Telecom & Technology USPTO Grants Google AI Image-Text Embedding Patent
Routine Notice Added Final

USPTO Grants Google AI Image-Text Embedding Patent

Favicon for changeflow.com USPTO Patent Grants - AI & Computing (G06N)
Published
Detected
Email

Summary

The USPTO granted patent US12608418B2 to Google LLC on April 21, 2026, covering methods for training image-text embedding models using training data that includes image pairs and selection data indicating co-click rates and similar-image click rates. The patent lists ten inventors including Zhen Li, Yi-Ting Chen, and Da-Cheng Juan.

Published by USPTO on changeflow.com . Detected, standardized, and enriched by GovPing. Review our methodology and editorial standards .

What changed

The USPTO issued patent US12608418B2 to Google LLC covering methods and systems for training image embedding models. The training data includes image pairs with selection data indicating co-click rates and similar-image click rates, which are used to train model parameters. The patent contains 20 claims.

Technology companies developing image search, visual search, or multimodal AI systems should consider reviewing this patent for prior art and freedom-to-operate purposes. The patent does not create immediate compliance obligations but represents part of the growing IP landscape around AI-driven image-text matching. No regulatory action is required.

Archived snapshot

Apr 21, 2026

GovPing captured this document from the original source. If the source has since changed or been removed, this is the text as it existed at that time.

← USPTO Patent Grants

Training image and text embedding models

Grant US12608418B2 Kind: B2 Apr 21, 2026

Assignee

GOOGLE LLC

Inventors

Zhen Li, Yi-Ting Chen, Yaxi Gao, Da-Cheng Juan, Aleksei Timofeev, Chun-Ta Lu, Futang Peng, Sujith Ravi, Andrew Tomkins, Thomas J. Duerig

Abstract

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for training an image embedding model. In one aspect, a method comprises: obtaining training data comprising a plurality of training examples, wherein each training example comprises: an image pair comprising a first image and a second image; and selection data indicating one or more of: (i) a co-click rate of the image pair, and (ii) a similar-image click rate of the image pair; and using the training data to train an image embedding model having a plurality of image embedding model parameters.

CPC Classifications

G06N 3/084 G06N 3/0427 G06N 3/0445 G06F 16/538 G06F 16/9538 G06F 16/55 G06K 9/6215 G06K 9/6254 G06K 9/6257

Filing Date

2024-06-12

Application No.

18741082

Claims

20

View original document →

Parties

Get daily alerts for USPTO Patent Grants - AI & Computing (G06N)

Daily digest delivered to your inbox.

Free. Unsubscribe anytime.

About this page

What is GovPing?

Every important government, regulator, and court update from around the world. One place. Real-time. Free. Our mission

What's from the agency?

Source document text, dates, docket IDs, and authority are extracted directly from USPTO.

What's AI-generated?

The summary, classification, recommended actions, deadlines, and penalty information are AI-generated from the original text and may contain errors. Always verify against the source document.

Last updated

Classification

Agency
USPTO
Published
April 21st, 2026
Instrument
Notice
Branch
Executive
Legal weight
Binding
Stage
Final
Change scope
Minor
Document ID
US12608418B2

Who this affects

Applies to
Technology companies
Industry sector
5112 Software & Technology
Activity scope
Patent grant IP registration
Geographic scope
United States US

Taxonomy

Primary area
Intellectual Property
Operational domain
Legal
Topics
Artificial Intelligence

Get alerts for this source

We'll email you when USPTO Patent Grants - AI & Computing (G06N) publishes new changes.

Free. Unsubscribe anytime.

You're subscribed!