Patents by Inventor Filip Gralinski

Filip Gralinski has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).

  • Iterative training for text-image-layout transformer
    Patent number: 11620451
    Abstract: Disclosed herein is a system and method for Natural Language Processing (NLP) of real world documents. the system and method combines various models not previously combined and overcomes the challenges of this combination. Models include an encoder-decoder model, a spatial model, and a multi-modal model. An iterative training process receives documents and generates outputs, wherein the iterative training process comprises enabling information retrieval from documents without training data.
    Type: Grant
    Filed: June 16, 2022
    Date of Patent: April 4, 2023
    Assignee: APPLICA SP. Z O.O.
    Inventors: Adam Dancewicz, Filip Gralinski, Lukasz Konrad Borchmann
  • ITERATIVE TRAINING FOR TEXT-IMAGE-LAYOUT TRANSFORMER
    Publication number: 20220327286
    Abstract: Disclosed herein is a system and method for Natural Language Processing (NLP) of real world documents. the system and method combines various models not previously combined and overcomes the challenges of this combination. Models include an encoder-decoder model, a spatial model, and a multi-modal model. An iterative training process receives documents and generates outputs, wherein the iterative training process comprises enabling information retrieval from documents without training data.
    Type: Application
    Filed: June 16, 2022
    Publication date: October 13, 2022
    Applicant: Applica sp. z o.o.
    Inventors: Adam Dancewicz, Filip Gralinski, Lukasz Konrad Borchmann
  • Iterative training for text-image-layout transformer
    Patent number: 11455468
    Abstract: Disclosed herein is a system and method for Natural Language Processing (NLP) of real world documents. the system and method combines various models not previously combined and overcomes the challenges of this combination. Models include an encoder-decoder model, a spatial model, and a multi-modal model. An iterative training process receives documents and generates outputs, wherein the iterative training process comprises enabling information retrieval from documents without training data.
    Type: Grant
    Filed: February 16, 2022
    Date of Patent: September 27, 2022
    Assignee: APPLICA SP. Z O.O.
    Inventors: Adam Dancewicz, Filip Gralinski, Lukasz Konrad Borchmann
  • ITERATIVE TRAINING FOR TEXT-IMAGE-LAYOUT TRANSFORMER
    Publication number: 20220261547
    Abstract: Disclosed herein is a system and method for Natural Language Processing (NLP) of real world documents. the system and method combines various models not previously combined and overcomes the challenges of this combination. Models include an encoder-decoder model, a spatial model, and a multi-modal model. An iterative training process receives documents and generates outputs, wherein the iterative training process comprises enabling information retrieval from documents without training data.
    Type: Application
    Filed: February 16, 2022
    Publication date: August 18, 2022
    Applicant: Applica sp. z o.o.
    Inventors: Adam Dancewicz, Filip Gralinski, Lukasz Konrad Borchmann