Generative Artificial Intelligence Extracts Structure-Function Relationships from Plants for New Materials
🤖 AI Summary
Designing humidity-responsive biomimetic materials remains challenging due to the complexity of translating biological principles into functional synthetic systems.
Method: We propose the first generative AI framework integrating plant science, biomimetics, and materials engineering. It employs a fine-tuned large language model—BioinspiredLLM—enhanced by retrieval-augmented generation (RAG), multi-agent collaborative reasoning, and hierarchical sampling to automatically extract structure–function relationships (e.g., hygroscopic actuation in pollen and *Rhapis excelsa* leaves) from cross-disciplinary literature and generate experimentally testable material design hypotheses.
Contribution/Results: The framework enables cross-domain knowledge transfer and structured hypothesis generation. Experimentally, we fabricated pollen-based adhesives with tunable morphology and shear strength, validating the feasibility and effectiveness of AI-generated designs. This work establishes a data-driven paradigm for biomimetic materials discovery.
Application Category
📝 Abstract
Large language models (LLMs) have reshaped the research landscape by enabling new approaches to knowledge retrieval and creative ideation. Yet their application in discipline-specific experimental science, particularly in highly multi-disciplinary domains like materials science, remains limited. We present a first-of-its-kind framework that integrates generative AI with literature from hitherto-unconnected fields such as plant science, biomimetics, and materials engineering to extract insights and design experiments for materials. We focus on humidity-responsive systems such as pollen-based materials and Rhapis excelsa (broadleaf lady palm) leaves, which exhibit self-actuation and adaptive performance. Using a suite of AI tools, including a fine-tuned model (BioinspiredLLM), Retrieval-Augmented Generation (RAG), agentic systems, and a Hierarchical Sampling strategy, we extract structure-property relationships and translate them into new classes of bioinspired materials. Structured inference protocols generate and evaluate hundreds of hypotheses from a single query, surfacing novel and experimentally tractable ideas. We validate our approach through real-world implementation: LLM-generated procedures, materials designs, and mechanical predictions were tested in the laboratory, culminating in the fabrication of a novel pollen-based adhesive with tunable morphology and measured shear strength, establishing a foundation for future plant-derived adhesive design. This work demonstrates how AI-assisted ideation can drive real-world materials design and enable effective human-AI collaboration.
Problem
Research questions and friction points this paper is trying to address.
Extracts structure-function relationships from plants for new materials
Integrates generative AI with plant science and materials engineering
Validates AI-generated designs through real-world material fabrication
Innovation
Methods, ideas, or system contributions that make the work stand out.
Integrates generative AI with multi-disciplinary literature
Uses BioinspiredLLM, RAG, agentic systems, Hierarchical Sampling
Validates AI-generated designs via real-world lab testing
🔎 Similar Papers
No similar papers found.
