Qumus: Realization of An Embodied AI Quantum Material Experimentalist

2026-05-18

AI Summary

This work proposes Qumus—the first AI quantum materials experimentalist—integrated into a robotic micro-laboratory to overcome the limitations of embodied artificial intelligence in real-world scientific discovery, particularly in integrating high-level reasoning, multimodal perception, and physical execution. Qumus employs a multi-agent architecture driven by large language models for high-order reasoning, coupled with multimodal sensing, closed-loop control, and autonomous error correction, enabling the first fully autonomous research cycle spanning hypothesis generation, experimental planning, atomic-scale manipulation, and result analysis. Demonstrating its capabilities, Qumus autonomously synthesized graphene and fabricated van der Waals stacked field-effect transistors, establishing a new paradigm for embodied AI in quantum materials discovery and validating its efficiency and generalizability.

Abstract

While modern Large Language Models (LLMs) and agentic artificial intelligence (AI) have demonstrated transformative capabilities in digital domains, the realization of embodied AI capable of real-world scientific discovery remains a difficult frontier. The advancements are hindered by the inherent complexity of integrating high-level reasoning, multimodal information processing and real-time physical execution. Here we introduce Qumus, the first AI quantum materials experimentalist. Physically embodied within a robotic mini-laboratory, Qumus is an intelligent, multimodal, and multi-agent system designed for the creation and nano-processing of atomically thin two-dimensional (2D) materials and stacked van der Waals (vdW) structures. Qumus autonomously navigates the full scientific cycle, from hypothesis generation and protocol planning to multi-step experimental execution, result analysis and reporting, acting as an experimentalist. Markedly, the system has achieved, for the first time, the AI-creation of graphene, as well as the first AI-fabrication of complex nanodevices including atomically thin field-effect transistors via vdW stacking. Qumus excels at these tasks by demonstrating autonomous error correction and closed-loop experimentation. Our results establish a generalizable framework for self-improving embodied AI systems that learn directly from the quantum world, opening a pathway toward accelerated discovery in quantum materials, electronics and beyond.

Problem

• embodied AI
• quantum materials
• autonomous experimentation
• scientific discovery
• robotic laboratory

Innovation

• embodied AI
• quantum materials
• autonomous experimentation
• van der Waals stacking
• robotic laboratory