LLEMA: Evolutionary Search with LLMs for Multi-Objective Materials Discovery

Nikhil Abhyankar, Sanchit Kabra, Saaketh Desai, Chandan K. Reddy

Keywords: Large Language Models, Evolutionary Optimization, AI for Science, Materials Discovery

TL;DR: LLM-guided Evolution for Materials design (LLEMA) is a unified evolutionary framework that couples the scientific knowledge embedded in large language models with chemistry-informed evolutionary rules and memory-based refinement.

Abstract: Materials discovery requires navigating vast chemical and structural spaces while satisfying multiple, often conflicting, objectives. We present LLM-guided Evolution for MAterials discovery (**LLEMA**), a unified framework that couples the scientific knowledge embedded in large language models with chemistry-informed evolutionary rules and memory-based refinement. At each iteration, an LLM proposes crystallographically specified candidates under explicit property constraints; a surrogate-augmented oracle estimates physicochemical properties; and a multi-objective scorer updates success/failure memories to guide subsequent generations. Evaluated on **14 realistic tasks** that span electronics, energy, coatings, optics, and aerospace, LLEMA discovers candidates that are chemically plausible, thermodynamically stable, and property-aligned, achieving higher hit rates and improved Pareto front quality relative to generative and LLM-only baselines. Ablation studies confirm the importance of rule-guided generation, memory-based refinement, and surrogate prediction. By enforcing synthesizability and multi-objective trade-offs, LLEMA provides a principled approach to accelerating practical materials discovery. Project website: https://scientific-discovery.github.io/llema-project/

Primary Area: foundation or frontier models, including LLMs

Submission Number: 23362