Xingchen Wan

Title: Innovations of Xingchen Wan

Introduction

Xingchen Wan is a notable inventor based in Oxford, GB. He has made significant contributions to the field of artificial intelligence, particularly in the development of large language models (LLMs). His innovative work focuses on enhancing the performance of LLMs through self-generated demonstrations.

Latest Patents

Xingchen Wan holds a patent titled "Self-improving LLMs through consistency-based self-generated demonstrations." This patent addresses the automatic selection of examples in a prompt for an LLM to demonstrate task performance. The disclosure outlines methods for selecting and building a set of examples from LLM zero-shot outputs based on predetermined criteria, which include consistency, diversity, and repetition. In experiments involving three different LLMs, his approach has shown to improve performance by up to 15% compared to zero-shot baselines and can match or exceed few-shot baselines for various reasoning tasks. He has 1 patent to his name.

Career Highlights

Xingchen Wan is currently employed at Google Inc., where he continues to push the boundaries of AI research. His work has garnered attention for its practical applications and innovative methodologies.

Collaborations

Xingchen collaborates with talented individuals in the field, including Ruoxi Sun and Hanjun Dai. Their combined expertise contributes to the advancement of AI technologies.

Conclusion

Xingchen Wan's contributions to the field of artificial intelligence, particularly through his patent on self-improving LLMs, highlight his innovative spirit and dedication to enhancing technology. His work at Google Inc. and collaborations with other experts further solidify his position as a leading inventor in the industry.