Ruoxi Sun

Title: Ruoxi Sun: Innovator in AI and Protein Structure Prediction

Introduction

Ruoxi Sun is a prominent inventor based in Mountain View, CA (US). She has made significant contributions to the fields of artificial intelligence and protein structure prediction. With a total of 3 patents, her work is at the forefront of technological advancements.

Latest Patents

One of her latest patents is titled "Self-improving LLMs through consistency-based self-generated demonstrations." This patent focuses on automatically selecting examples in a prompt for a large language model (LLM) to demonstrate task performance. The innovation allows for the selection and construction of a set of examples from LLM zero-shot outputs based on predetermined criteria, which can enhance consistency, diversity, and repetition. In tests involving three different LLMs, her methods improved performance by up to 15% compared to zero-shot baselines and matched or exceeded few-shot baselines for various reasoning tasks.

Another notable patent is "Predicting protein structures using geometry neural networks that estimate similarity between predicted protein structures and actual protein structures." This invention encompasses methods, systems, and apparatus for protein structure prediction. It involves determining alternative predicted structures of proteins and processing them using a geometry neural network to generate an output that estimates the similarity between predicted and actual protein structures.

Career Highlights

Ruoxi Sun has worked with leading companies in the tech industry, including DeepMind Technologies Limited and GDM Holding LLC. Her experience in these organizations has allowed her to develop and refine her innovative ideas.

Collaborations

Throughout her career, Ruoxi has collaborated with notable professionals, including Andrew W. Senior and James Kirkpatrick. These collaborations have contributed to her success and the advancement of her research.

Conclusion

Ruoxi Sun is a trailblazer in the fields of artificial intelligence and protein structure prediction. Her innovative patents and collaborations highlight her significant impact on technology and science.