Jpmorgan Chase Bank, N.a.

Simerjot Kaur

Sameena Shah

Armineh Nourbakhsh

Daniel Borrajo

Mathieu Sibue

Akshat Gupta

Charese Smiley

Suchetha Siddagangappa

Manuela Veloso

Natraj Raman

Russell Kociuba

Keshav Ramani

Cecilia Tilli

Andrea Stefanucci

Yulong Pei

Petr Babkin

Joy Sain

Jade Fallen

Elena Kochkina

Pietro Totis

Dongsheng Wang

Samuel Angmor Mensah

Amit Aswani

Toyin Aguda

Brian Bramble

Xiaomo Liu

Alex Woodgate

Steven Pomerville

Zhiqiang Ma

Jonathan Roger Horn

Various methods and processes, apparatuses/systems, and media for performing spatial-aware reading for visual documents are disclosed. A processor implements a language model; modifies existing parameters and architecture of the language model to incorporate new parameters in its architecture by integrating a disentangled spatial attention process to the language model; pretrains the modified model by performing an autoregressive block infilling process on a plurality of document pages thereby training the new parameters and further adjusting the existing parameters; instruction-tunes the pretrained model on data derived from a plurality of visually rich document understanding datasets to teach the pretrained model to follow document-oriented instructions or answer questions about documents by leveraging their content and their spatial layout and outputting a trained model; and performs spatial-aware reading for visual documents by utilizing the trained model.

System and method for layout-aware generative pretraining for visually rich document understanding

A method for automating a process of generating messages that are responsive to client inquiries by using an AI algorithm that implements a machine learning technique to ensure accuracy and timeliness in the responses is provided. The method includes: receiving a first message that includes an inquiry that relates to an account associated with a user; applying an AI algorithm for analyzing the first message in order to extract information that relates to the inquiry; determining, based on a result of the analysis, whether generating a response to the inquiry requires human intervention; when human intervention is not required, retrieving information that is responsive to the inquiry from a memory; generating a second message that includes the information that is responsive to the inquiry; and transmitting the second message to the user.

Method and system for automated message generation

A method and a system for automatically processing financial documents to generate knowledge graphs that convey information relating to entities of interest and relationships between those entities are provided. The method includes: receiving a document; extracting raw text included in the document; identifying, based on the extracted raw text, a set of entities that are named in the document; determining respective relationship information that corresponds to respective pairs of entities; constructing a knowledge graph that illustrates respective relationships among the respective pairs of entities; and outputting the knowledge graph. The determination of the respective relationship information may be performed by applying an artificial intelligence (AI) algorithm that is trained by using historical data that relates to the set of entities.

Method and system for understanding financial documents

A method and a system for extracting information from a subpoena document are provided. The method includes: receiving a subpoena document; extracting raw text included in the subpoena document; identifying, based on the extracted raw text, entities that are named in the subpoena document; determining, based on the extracted raw text, first information that relates to a scope period, a law enforcement agency, and/or an investigative agent associated with the subpoena document; retrieving second information that relates to the identified entities from a customer database; and outputting a subset of the determined first information and a subset of the obtained second information. The method may also include using a weighted fuzzy name match algorithm to match the identified entities with the second information.

Method and system for processing subpoena documents

A method for using an artificial intelligence-based framework to automatically discover and extract information about private and public companies that is accessible from public data sources is provided. The method includes: receiving first information that relates to an identification of a company; determining, based on the first information, publicly accessible data sources via which second information that relates to the company is available; receiving user input that relates to a type of company-specific data to be accessed from the data sources; retrieving, based on the received user input, a subset of the second information; and outputting the subset of the second information. The second information may include leadership profiles, information about products and services, a company description, and/or information about corporate policies.

Method and system for automated public information discovery

Growing community of inventors

Jersey City, NJ, United States of America

Simerjot Kaur