Max Robotics — Global Service Robot Directory + US Certification

About

Sergio Arnaud is a leading researcher at the intersection of embodied AI, robotics, and self-supervised learning. His work focuses on enabling machines to understand, navigate, and act within physical environments through minimal supervision. Arnaud’s most influential contribution is **OpenEQA** (46 citations), which redefines Embodied Question Answering for the foundation model era—allowing agents on smart glasses or robots to answer natural language questions by leveraging episodic memory. He also developed **Adaptive Skill Coordination (ASC)** (36 citations), a modular framework for long-horizon mobile manipulation tasks like pick-and-place, combining visuomotor skills with high-level planning. Most recently, his work on **V-JEPA 2** (2025) explores how self-supervised video models, trained on internet-scale data with minimal robot interaction, can achieve world understanding, prediction, and planning—a step toward generalist agents. With over 85 citations across his top papers, Arnaud is shaping the future of embodied intelligence, bridging perception, reasoning, and physical action. His research is essential reading for anyone interested in how AI can learn to see, understand, and move through the world.