Yangyi Chen

Title: Innovations of Yangyi Chen in Visual Language Models

Introduction

Yangyi Chen is an accomplished inventor based in Princeton, NJ (US). He has made significant contributions to the field of visual language models, particularly through his innovative patent. His work focuses on enhancing the interaction between visual queries and language processing, which is crucial for advancing artificial intelligence applications.

Latest Patents

Yangyi Chen holds a patent titled "Instruction-guided visual embeddings and feedback-based learning in large vision-language models." This patent describes a method for fine-tuning a Large Visual Language Model (LVLM). The process involves providing visual queries that consist of images and related textual queries. The LVLM processes these queries to extract visual embeddings, generating responses based on the corresponding visual embeddings. The method also includes evaluating the generated responses to ensure they meet predefined criteria and providing feedback for further fine-tuning of the LVLM.

Career Highlights

Yangyi Chen is currently employed at SRI International, where he continues to push the boundaries of research in visual language models. His work has implications for various applications, including image recognition and natural language processing. With a focus on integrating visual and textual information, he is at the forefront of innovation in this rapidly evolving field.

Collaborations

Yangyi collaborates with notable colleagues such as Karan Sikka and Michael A. Cogswell. Their combined expertise contributes to the advancement of research and development in visual language technologies.

Conclusion

Yangyi Chen's contributions to the field of visual language models exemplify the innovative spirit of modern inventors. His patent and ongoing work at SRI International highlight the importance of integrating visual and textual data in artificial intelligence.