Structural Constraint Integration in Generative Model for Discovery of Quantum Material Candidates
This work addresses the problem of discovering new quantum materials for the materials science community by enabling the generation of stable, geometrically constrained compounds, though it is incremental as it builds on existing diffusion models.
The authors tackled the challenge of integrating geometric patterns into generative models for discovering quantum materials by introducing SCIGEN, a method that modifies diffusion models to steer generation toward constrained outputs, resulting in over 10% of 8 million generated compounds surviving stability pre-screening and over 50% of 26,000 compounds passing DFT-level structural optimization.
Billions of organic molecules are known, but only a tiny fraction of the functional inorganic materials have been discovered, a particularly relevant problem to the community searching for new quantum materials. Recent advancements in machine-learning-based generative models, particularly diffusion models, show great promise for generating new, stable materials. However, integrating geometric patterns into materials generation remains a challenge. Here, we introduce Structural Constraint Integration in the GENerative model (SCIGEN). Our approach can modify any trained generative diffusion model by strategic masking of the denoised structure with a diffused constrained structure prior to each diffusion step to steer the generation toward constrained outputs. Furthermore, we mathematically prove that SCIGEN effectively performs conditional sampling from the original distribution, which is crucial for generating stable constrained materials. We generate eight million compounds using Archimedean lattices as prototype constraints, with over 10% surviving a multi-staged stability pre-screening. High-throughput density functional theory (DFT) on 26,000 survived compounds shows that over 50% passed structural optimization at the DFT level. Since the properties of quantum materials are closely related to geometric patterns, our results indicate that SCIGEN provides a general framework for generating quantum materials candidates.