A Chemically Grounded Evaluation Framework for Generative Models in Materials Discovery

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Veillon, Elohan, Klipfel, Astrid, Sayede, Adlane, Bouraoui, Zied
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911350906683392
author Veillon, Elohan
Klipfel, Astrid
Sayede, Adlane
Bouraoui, Zied
author_facet Veillon, Elohan
Klipfel, Astrid
Sayede, Adlane
Bouraoui, Zied
contents Generative models hold great promise for accelerating materials discovery, but their evaluation often overlooks the chemical validity and stability requirements crucial to real-world applications. Density Functional Theory (DFT) simulations are the gold standard for evaluating such properties but are computationally intensive and inaccessible to non-experts. We propose a chemically grounded, user-friendly evaluation framework that integrates DFT-based stability analysis with commonly used machine learning (ML) metrics. Through systematic experiments using both perturbative and generative methods, we demonstrate that conventional ML metrics can misrepresent chemical feasibility. To address this, we propose new insights on robust metrics and highlight the importance of simulation-informed evaluation for developing reliable generative models in materials science.
format Preprint
id arxiv_https___arxiv_org_abs_2601_00886
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle A Chemically Grounded Evaluation Framework for Generative Models in Materials Discovery
Veillon, Elohan
Klipfel, Astrid
Sayede, Adlane
Bouraoui, Zied
Materials Science
Generative models hold great promise for accelerating materials discovery, but their evaluation often overlooks the chemical validity and stability requirements crucial to real-world applications. Density Functional Theory (DFT) simulations are the gold standard for evaluating such properties but are computationally intensive and inaccessible to non-experts. We propose a chemically grounded, user-friendly evaluation framework that integrates DFT-based stability analysis with commonly used machine learning (ML) metrics. Through systematic experiments using both perturbative and generative methods, we demonstrate that conventional ML metrics can misrepresent chemical feasibility. To address this, we propose new insights on robust metrics and highlight the importance of simulation-informed evaluation for developing reliable generative models in materials science.
title A Chemically Grounded Evaluation Framework for Generative Models in Materials Discovery
topic Materials Science
url https://arxiv.org/abs/2601.00886