Zekai Wang

About

Zekai Wang is a pioneering researcher at the intersection of large language models (LLMs) and robotics, with a primary focus on enabling embodied intelligence through in-context learning. His most notable contribution is the development of RoboPrompt, a framework that harnesses LLMs’ in-context learning abilities to directly predict robot actions—a novel approach that bridges the gap between language understanding and physical task execution. This work, published in 2025, has already garnered 3 citations, signaling early impact in a rapidly evolving field. Wang’s research addresses a critical challenge: while LLMs excel in language tasks, their potential for zero-shot or few-shot robot action prediction was largely untapped before his work. By demonstrating that LLMs can generalize from a handful of examples to generate precise motor commands, he opens new pathways for more adaptable and intuitive human-robot collaboration. His achievements are particularly relevant for students and researchers exploring how foundation models can be repurposed for real-world control, making him a rising figure in the movement toward generalist robots that learn on the fly.

Research Focus

Key Achievements

Top Papers

1. 1
   In-Context Learning Enables Robot Action Prediction in LLMs

   3 citations · 2025

Key Collaborators

Contact & Links