The patent badge is an abbreviated version of the USPTO patent document. The patent badge does contain a link to the full patent document.

The patent badge is an abbreviated version of the USPTO patent document. The patent badge covers the following: Patent number, Date patent was issued, Date patent was filed, Title of the patent, Applicant, Inventor, Assignee, Attorney firm, Primary examiner, Assistant examiner, CPCs, and Abstract. The patent badge does contain a link to the full patent document (in Adobe Acrobat format, aka pdf). To download or print any patent click here.

Date of Patent:
Jul. 28, 2026

Filed:

Jan. 11, 2024
Applicant:

Intuit Inc., Mountain View, CA (US);

Inventors:

Benjamin Adetor Kwapong, San Jose, CA (US);

Hui Chen, San Jose, CA (US);

Tharathorn Joy Rimchala, San Francisco, CA (US);

Assignee:

Intuit Inc., Mountain View, CA (US);

Attorney:
Primary Examiner:
Int. Cl.
CPC ...
G06V 30/19 (2022.01); G06V 30/412 (2022.01); G06V 30/414 (2022.01);
U.S. Cl.
CPC ...
G06V 30/19147 (2022.01); G06V 30/412 (2022.01); G06V 30/414 (2022.01);
Abstract

Systems and methods are disclosed for training an electronic document extraction model, including the generation of the training data to train the model based on sampling a pool of electronic documents based on a rareness metric of the documents. Each electronic document has a document-level rareness metric generated, with the document-level rareness metric being based on one or more of a structural rareness metric or a content rareness metric of the document. The structural rareness metric measures the rareness of the document structure, which may be irrespective of the text content of the document. The content rareness metric measures the rareness of the document content, which may be irrespective of the document structure. The electronic documents are sampled based on the document-level rareness metrics to increase the number of rare documents in the training data without unduly biasing the sampling to optimize the training data for training the extraction model.


Find Patent Forward Citations

Kwapong, Benjamin. (2026). Training of an electronic document extraction model (U.S. Patent No. 12694702). U.S. Patent and Trademark Office. https://idiyas.com/patent/badge/12694702

Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile

Loading…