Authors use an algorithm to create 20 millions potential new materials and go through a bunch of test to check for stability. They end up with 142 compounds...
They proceed to check the potential properties for some of these new materials.

I like that the starting point relies on crystallographically-sensible reasoning and clearly relies on an understanding of the science at play and not just "let throw everything in a big computer".

https://arxiv.org/pdf/2508.04110
#crystallography #MaterialScience

**CCochard** @CCochard@sciences.re · Aug 6 *

Aug 6 *

CCochard @CCochard@sciences.re

"Artificial Intelligence and Generative Models for Materials Discovery: A Review"

http://arxiv.org/abs/2508.03278

Really not my speciality but keeping an eye on AI for material sciences to have some opinion on the topic. And you can see that the paper was written by people with not-so-deep knowledge of material science (but a lot better than other works I have seen...)
Not against opinion on people more skilled than me!

A thread
#MaterialScience #AI

arXiv.orgArtificial Intelligence and Generative Models for Materials Discovery -- A ReviewHigh throughput experimentation tools, machine learning (ML) methods, and open material databases are radically changing the way new materials are discovered. From the experimentally driven approach in the past, we are moving quickly towards the artificial intelligence (AI) driven approach, realizing the 'inverse design' capabilities that allow the discovery of new materials given the desired properties. This review aims to discuss different principles of AI-driven generative models that are applicable for materials discovery, including different materials representations available for this purpose. We will also highlight specific applications of generative models in designing new catalysts, semiconductors, polymers, or crystals while addressing challenges such as data scarcity, computational cost, interpretability, synthesizability, and dataset biases. Emerging approaches to overcome limitations and integrate AI with experimental workflows will be discussed, including multimodal models, physics informed architectures, and closed-loop discovery systems. This review aims to provide insights for researchers aiming to harness AI's transformative potential in accelerating materials discovery for sustainability, healthcare, and energy innovation.