Sivan Doveh

Title: Sivan Doveh: Innovator in Visual Language Grounding Models

Introduction

Sivan Doveh is a prominent inventor based in Ramat Gan, Israel. She has made significant contributions to the field of visual language grounding, showcasing her expertise through innovative patent work.

Latest Patents

Sivan holds a patent titled "Training visual language grounding models using separation loss." This patent describes a system that includes a processor designed to receive a randomly generated alpha-map, a pair of training images, and a pair of training texts associated with those images. The processor generates a blended image based on the alpha-map and the training images. It then trains a visual language grounding model to separate the blended image into heatmaps that identify portions corresponding to each training image using a separation loss. This innovative approach enhances the understanding of visual language models.

Career Highlights

Sivan is currently employed at International Business Machines Corporation, commonly known as IBM. Her work at IBM allows her to collaborate with leading experts in the field and contribute to cutting-edge research and development.

Collaborations

Sivan has worked alongside notable colleagues, including Assaf Arbelle and Leonid Karlinsky. These collaborations have further enriched her research and innovation efforts.

Conclusion

Sivan Doveh is a trailblazer in the realm of visual language grounding, with her patent work reflecting her innovative spirit and dedication to advancing technology. Her contributions are paving the way for future developments in this exciting field.