This inventor holds 1 USPTO granted patent. Top assignee: Google 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: Raphael Dominik Hoffman: Innovator in Instruction Fine-tuning Datasets
Introduction
Raphael Dominik Hoffman is an inventor based in Los Altos, CA (US). He is known for his work in developing systems and methods for generating instruction fine-tuning datasets for general-purpose embedding models. His innovative approach aims to enhance the efficiency and effectiveness of machine learning models.
Latest Patent Applications
Hoffman's notable patent application is titled "Systems and Methods for Generating Instruction Fine-tuning Dataset for a General Purpose Embedding Model." This application outlines a method that includes providing a sequence model with a plurality of few-shot prompts. Each prompt consists of a demonstration passage, a demonstration task, and a demonstration query. The demonstration task describes a type of retrieval, while the demonstration query is relevant to the task. The method also involves receiving a plurality of predicted task-query pairs from the sequence model based on the input passage and the few-shot prompts. Furthermore, it includes generating a synthetic training dataset that comprises the plurality of passages and the respective predicted task-query pairs.
Conclusion
Raphael Dominik Hoffman is an emerging inventor whose work focuses on advancing machine learning through innovative dataset generation methods. His contributions have the potential to significantly impact the field of artificial intelligence.
Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile
