From Generative Models to Real-World Materials: Entalpic’s AI-Driven Discovery Loop

Alexandre Duval

Keywords: materials discovery, generative models, MLIPs, active learning, LeMaterial, high-throughput lab experiments

TL;DR: We present Entalpic’s AI-driven discovery platform that integrates LeMaterial quantum datasets, predictive GNNs, generative GFlowNets, agentic LLMs, and experimental feedback into a closed-loop system for materials innovation.

Confirmation Of Submission Requirements: I submit an abstract. It uses the template provided on the submission page and is no longer than 2 pages.

Submission Number: 298