This inventor holds 1 USPTO granted patent. Top assignee: Gdm Holding LLC. 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: Fabian Julius Mentzer: Inventor in Multimodal Data Generation
Introduction
Fabian Julius Mentzer is an inventor based in Zurich, Switzerland. His work includes a patent on generating multimodal data with neural networks.
Latest Patents
Mentzer is associated with 1 patent: “Generating Multimodal Data Using a Media Item Generation Neural Network and a Token Generation Neural Network.” The patent describes a computer-implemented method in which a token generation neural network autoregressively produces a sequence of multimodal tokens. A media item, such as an image, audio, or video, can be generated based on features of the current sequence and an encoded representation of input media produced by a media encoder neural network.
Career Highlights
Mentzer is associated with GDM Holding LLC. His patent focuses on neural-network methods for generating multimodal content conditioned on both generated tokens and input media.
Collaborations
The provided information identifies Kaushik Shivakumar and Robert Junior Riachi as Mentzer’s coworkers.
Conclusion
Fabian Julius Mentzer’s documented patent work centers on neural-network techniques for generating multimodal data, including media conditioned on input content and token sequences.
Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile