LacMaterial: Large Language Models as Analogical Chemists for Materials Discovery

2025-10-25

AI Summary

Human scientific discovery is often constrained by domain-specific knowledge limitations and surface-level similarity biases, hindering cross-disciplinary, structure-driven analogical reasoning and impeding innovation in novel materials. Method: We propose an explicit analogical reasoning framework powered by large language models (LLMs). It first performs cross-domain analogical retrieval to transcend disciplinary boundaries, then constructs in-domain structural analogy templates from few-shot labeled examples, integrating relation extraction, knowledge transfer, and prompt engineering to generate interpretable hypotheses. Contribution/Results: The framework shifts battery material design from empirical dopant substitution toward deep structural-guided innovation. In candidate material generation, it significantly outperforms standard prompting baselines and successfully identifies multiple electrochemically promising novel materials—providing the first empirical validation of LLMs as capable “analogical chemists.”

Technology Category

• Cognitive Modeling & Cognitive Systems: Analogy
• Machine Learning: Large Multimodal Models (LMMs)
• Knowledge Representation and Reasoning: Computational Complexity of Reasoning

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Economics, Online Markets and Human Computation: Uses of LLMs and GenAI for marketplace design, bidding, and strategic interactions

Abstract

Analogical reasoning, the transfer of relational structures across contexts (e.g., planet is to sun as electron is to nucleus), is fundamental to scientific discovery. Yet human insight is often constrained by domain expertise and surface-level biases, limiting access to deeper, structure-driven analogies both within and across disciplines. Large language models (LLMs), trained on vast cross-domain data, present a promising yet underexplored tool for analogical reasoning in science. Here, we demonstrate that LLMs can generate novel battery materials by (1) retrieving cross-domain analogs and analogy-guided exemplars to steer exploration beyond conventional dopant substitutions, and (2) constructing in-domain analogical templates from few labeled examples to guide targeted exploitation. These explicit analogical reasoning strategies yield candidates outside established compositional spaces and outperform standard prompting baselines. Our findings position LLMs as interpretable, expert-like hypothesis generators that leverage analogy-driven generalization for scientific innovation.

Problem

• LLMs generate novel battery materials via cross-domain analogies
• Overcome human expertise limitations in analogical reasoning
• Construct in-domain analogical templates from few examples

Innovation

• Retrieving cross-domain analogs to steer material exploration
• Constructing in-domain analogical templates from few examples
• Leveraging analogy-driven generalization for scientific innovation