Patents by Inventor Thomas Hartvigsen

Thomas Hartvigsen 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).

  • ADVERSARIAL LANGUAGE IMITATION WITH CONSTRAINED EXEMPLARS
    Publication number: 20240339111
    Abstract: Generally discussed herein are devices, systems, and methods for generating a phrase that is confusing to a language classifier. A method can include determining, by the LC, a first classification score (CS) of a prompt indicating whether the prompt is a first class or a second class, predicting, based on the prompt and by a pre-trained language model (PLM), likely next words and a corresponding probability for each of the likely next words, determining, by the LC, a second CS for each of the likely next words, determining, by an adversarial classifier, respective scores for each of the likely next words, the respective scores determined based on the first CS of the prompt, the second CS of the likely next words, and the probabilities of the likely next words, and selecting, by an adversarial classifier, a next word of the likely next words based on the respective scores.
    Type: Application
    Filed: June 11, 2024
    Publication date: October 10, 2024
    Inventors: Hamid PALANGI, Saadia Kai GABRIEL, Thomas HARTVIGSEN, Dipankar RAY, Semiha Ece KAMAR EDEN
  • Adversarial language imitation with constrained exemplars
    Patent number: 12033617
    Abstract: Generally discussed herein are devices, systems, and methods for generating a phrase that is confusing to a language classifier. A method can include determining, by the LC, a first classification score (CS) of a prompt indicating whether the prompt is a first class or a second class, predicting, based on the prompt and by a pre-trained language model (PLM), likely next words and a corresponding probability for each of the likely next words, determining, by the LC, a second CS for each of the likely next words, determining, by an adversarial classifier, respective scores for each of the likely next words, the respective scores determined based on the first CS of the prompt, the second CS of the likely next words, and the probabilities of the likely next words, and selecting, by an adversarial classifier, a next word of the likely next words based on the respective scores.
    Type: Grant
    Filed: February 17, 2022
    Date of Patent: July 9, 2024
    Assignee: Microsoft Technology Licensing, LLC
    Inventors: Hamid Palangi, Saadia Kai Gabriel, Thomas Hartvigsen, Dipankar Ray, Semiha Ece Kamar Eden
  • ADVERSARIAL LANGUAGE IMITATION WITH CONSTRAINED EXEMPLARS
    Publication number: 20230260506
    Abstract: Generally discussed herein are devices, systems, and methods for generating a phrase that is confusing to a language classifier. A method can include determining, by the LC, a first classification score (CS) of a prompt indicating whether the prompt is a first class or a second class, predicting, based on the prompt and by a pre-trained language model (PLM), likely next words and a corresponding probability for each of the likely next words, determining, by the LC, a second CS for each of the likely next words, determining, by an adversarial classifier, respective scores for each of the likely next words, the respective scores determined based on the first CS of the prompt, the second CS of the likely next words, and the probabilities of the likely next words, and selecting, by an adversarial classifier, a next word of the likely next words based on the respective scores.
    Type: Application
    Filed: February 17, 2022
    Publication date: August 17, 2023
    Inventors: Hamid Palangi, Saadia Kai Gabriel, Thomas Hartvigsen, Dipankar Ray, Semiha Ece Kamar Eden