Leveraging Language Representation for Material Recommendation, Ranking, and Exploration
This work addresses the bottleneck of material discovery for researchers in materials science by providing a task-agnostic framework, though it is incremental as it builds on existing language model techniques.
The authors tackled the problem of limited general representations for exploring the material search space by introducing a framework that uses natural language embeddings from language models to represent compositional and structural features, resulting in diversified recommendations of prototype structures and identification of under-studied high-performance material spaces, with novel materials corroborated by first-principles calculations and experiments.
Data-driven approaches for material discovery and design have been accelerated by emerging efforts in machine learning. However, general representations of crystals to explore the vast material search space remain limited. We introduce a material discovery framework that uses natural language embeddings derived from language models as representations of compositional and structural features. The discovery framework consists of a joint scheme that first recalls relevant candidates, and next ranks the candidates based on multiple target properties. The contextual knowledge encoded in language representations conveys information about material properties and structures, enabling both representational similarity analysis for recall, and multi-task learning to share information across related properties. By applying the framework to thermoelectrics, we demonstrate diversified recommendations of prototype structures and identify under-studied high-performance material spaces. The recommended materials are corroborated by first-principles calculations and experiments, revealing novel materials with potential high performance. Our framework provides a task-agnostic means for effective material recommendation and can be applied to various material systems.