Inverse Materials Design by Large Language Model-Assisted Generative Framework

2025-02-25

AI Summary

To address low generation efficiency and high bias arising from data scarcity in inverse materials design, this work proposes AlloyGAN—the first closed-loop generative framework integrating large language models (LLMs) with conditional generative adversarial networks (cGANs). AlloyGAN automatically extracts structured prior knowledge from scientific literature via LLM-driven information mining, guiding cGANs to jointly generate alloy compositions and target properties under thermodynamic constraints. It further incorporates a thermodynamic property prediction module and experimental validation feedback, establishing a full-cycle pipeline: “knowledge mining → conditional generation → property prediction → experimental iteration.” In metallic glass design, AlloyGAN achieves <8% error in thermodynamic property prediction, attains a 73% experimental validation rate for top candidates, and reduces design cycle time by several-fold. The framework significantly enhances accuracy, interpretability, and cross-task transferability in few-shot inverse design scenarios.

Technology Category

• Natural Language Processing: Generation
• Machine Learning: Large Multimodal Models (LMMs)
• Humans and AI: Game Design — Procedural Content Generation & Storytelling

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications
• Social Networks and Social Media: Generative AI / large language models and their impact on social systems

Abstract

Deep generative models hold great promise for inverse materials design, yet their efficiency and accuracy remain constrained by data scarcity and model architecture. Here, we introduce AlloyGAN, a closed-loop framework that integrates Large Language Model (LLM)-assisted text mining with Conditional Generative Adversarial Networks (CGANs) to enhance data diversity and improve inverse design. Taking alloy discovery as a case study, AlloyGAN systematically refines material candidates through iterative screening and experimental validation. For metallic glasses, the framework predicts thermodynamic properties with discrepancies of less than 8% from experiments, demonstrating its robustness. By bridging generative AI with domain knowledge and validation workflows, AlloyGAN offers a scalable approach to accelerate the discovery of materials with tailored properties, paving the way for broader applications in materials science.

Problem

• Enhance inverse materials design efficiency
• Improve accuracy with LLM-assisted framework
• Accelerate discovery of tailored properties materials

Innovation

• LLM-assisted text mining
• Conditional Generative Adversarial Networks
• Iterative screening and validation