New York City, NY, United States of America

Yinan Zhai

This inventor holds 1 USPTO granted patent. Top assignee: Citibank, N.a.. Active years: 2026.

IDiyas Innovation Intelligence. (2026). Inventor Profile: Yinan Zhai. Retrieved from https://idiyas.com/inventor/yinan-zhai

Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile

Calculating rank

Average Co-Inventor Count = 1.0


Company Filing History:


Years Active: 2026

Loading Chart...
1 patent (USPTO):Explore Patents

Title: Yinan Zhai and Hierarchical Multi-Level Model Distillation.

Introduction

Yinan Zhai is an inventor based in New York City, New York, and is associated with Citibank, N.A. His patent work addresses ways to adapt large language models into specialized artificial intelligence systems for tasks with different categories and computational demands.

Latest Patents

Zhai is associated with 1 patent: “Hierarchical Multi-Level Model Distillation.” The disclosure describes determining a hierarchy of specialized generative models based on task categories and complexity levels. A large language model trains the specialized models to respond to category-related requests, while pruning and quantization adapt them to different levels of complexity and precision. The system also determines suitable deployment locations and transmits the models for use across diverse computational environments.

Career Highlights

The patent reflects an approach to deploying artificial intelligence models with varied capabilities and resource requirements. By organizing models into levels, the disclosed system aims to match responses and computational demands to the needs of particular tasks.

Collaborations

Yinan Zhai’s listed coworker is Ganesh Prasad Bhat. The available information identifies their professional association but does not specify their roles in developing the patent.

Conclusion

Yinan Zhai’s patent focuses on creating, adapting, and deploying specialized models derived from large language models, with the goal of supporting tasks across different complexity levels.

Profile summary based on public USPTO records.
Data Sources: USPTO Patent Grant XML, Patent Center, EPO & CIPO • Normalized by IDiyas Innovation Graph. Methodology & provenance architecturePlease report any incorrect information to [email protected]
Loading…