This inventor holds 3 USPTO granted patents and 12 published patent applications. Top assignee: Dk Crown Holdings Inc.. Active years: 2026.
Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile
Company Filing History:
Years Active: 2026
Title: Gregory Von Pless: Patents in Language-Model Systems
Introduction
Gregory Von Pless is an inventor based in Melrose, Massachusetts. His patent work includes systems that use language models to process profile data and route prompts in distributed networking environments. He is associated with Dk Crown Holdings Inc. and holds 3 patents.
Latest Patents
“Systems and Methods for Processing User Profile Data Structures Using Language Model Output” describes interactive systems that use language-model output to identify data structures relevant to a request and update a user profile. The system maintains data structures associated with events, processes prompts, generates a command based on selected data, and updates the profile accordingly.
“Systems and Methods for Routing Machine-Learning Prompts in a Distributed Networking Environment” describes a system that classifies the intent of a prompt and selects a suitable language model from multiple models. The selected model generates a text response, with the routing process designed to support communication sessions in a distributed environment.
Career Highlights
Von Pless’s listed patents focus on applying language models to practical data-processing and communication tasks. One invention addresses profile updates using structured information, while the other concerns matching prompts with models according to intent.
Collaborations
His listed coworkers include Robin Mohseni and Gengyuan Zhang. Their names appear alongside Von Pless’s professional and patent-related information.
Conclusion
Gregory Von Pless’s patent portfolio highlights innovations in language-model-driven profile management and prompt routing, reflecting a focus on applying machine learning to interactive systems.
Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile