logo

For the Inventor, By the Inventor

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.

Patent No.:

US 12717834 B1

PDFFull Text
Date of Patent:
Aug. 25, 2026

Filed:

Jan. 16, 2025

Systems and methods for generative ai-enhanced localized small language model data processing

Applicant:

Bank of America Corporation, Charlotte, NC (US);

Inventors:

Saurabh Arora, Gurugram, IN;

Sandeep Kumar Chauhan, Hyderabad, IN;

Amit Kumar Kolli, Hyderabad, CN;

Assignee:

BANK OF AMERICA CORPORATION, Charlotte, NC (US);

Attorneys:

Moore & Van Allen PLLC

Vincent J. Dowling

Primary Examiner:

Srirama Channavajjala

Int. Cl.
CPC ...
G06F 16/00 (2019.01); G06F 16/2455 (2019.01); G06F 16/34 (2019.01); G06F 16/9535 (2019.01); G06F 16/9538 (2019.01); G06F 40/00 (2020.01); G06F 40/117 (2020.01); G06F 40/284 (2020.01); G06N 20/00 (2019.01); G06V 30/12 (2022.01); G06V 30/413 (2022.01);
U.S. Cl.
CPC ...
G06F 16/345 (2019.01); G06F 16/24556 (2019.01); G06F 16/9535 (2019.01); G06F 16/9538 (2019.01); G06F 40/117 (2020.01); G06F 40/284 (2020.01); G06N 20/00 (2019.01); G06V 30/133 (2022.01); G06V 30/413 (2022.01);
Abstract

Systems, computer program products, and methods are described herein for generative AI-enhanced localized small language model data processing. Embodiments of the present disclosure are configured to receive a plurality of small language models, where each small language model is configured to process a file type of a plurality of file types, receive a datafile including one or more data entries, where each data entry has a data type of one or more data types, classify the datafile as being a file type, ingest the datafile in a small language model, where the small language model is configured to process the file type of the datafile, generate a data summarization of the one or more data entries, validate the data summarization via one or more additional small language models configured to process other file types of the plurality of file types, and output the validated data summarization.

Find Patent Forward Citations

Loading…