Max Robotics — Global Service Robot Directory + US Certification

About

Wenlong Huang is a leading researcher at the intersection of large language models (LLMs) and embodied robotics, pioneering new paradigms for grounding language in physical action. His work fundamentally reimagines how robots can leverage LLMs not just for planning, but for direct, code-driven control. Huang’s most influential contribution is the "Code as Policies" framework (561 citations), which repurposes code-writing LLMs to synthesize robot policy code from natural language commands, effectively turning language into executable robot behavior. He further advanced this vision with "PaLM-E" (350 citations), an embodied multimodal language model that directly incorporates real-world sensor data into LLMs, enabling grounded reasoning. His "Inner Monologue" (206 citations) demonstrated how LLMs can perform closed-loop planning by incorporating feedback from the environment, while "VoxPoser" (87 citations) introduced composable 3D value maps for dexterous manipulation without pre-defined motion primitives. Huang’s work has been recognized for its novelty and impact, receiving widespread attention at top robotics and AI venues. His research is shaping a future where robots understand and act upon human language with unprecedented flexibility, making him a pivotal figure in the emerging field of LLM-driven robotics.