Enhancing Large Language Models with Domain-Specific Knowledge: The Case in Topological Materials

2024-09-10

AI Summary

To address the limitations of large language models (LLMs) in domain-specific expertise on topological materials and the prohibitive computational cost of full-parameter fine-tuning, this work introduces MaterialsKG—the first domain-specific knowledge graph for topological quantum matter—constructed by semantically integrating extensive scientific literature. We propose a knowledge graph-enhanced, context-aware prompt learning framework that enables lightweight, scalable, and parameter-efficient adaptation of LLMs without full-parameter fine-tuning. Our approach innovatively couples structured domain knowledge with advanced prompt engineering. Experimental results demonstrate substantial improvements across key tasks: question-answering accuracy increases by 42% over general-purpose LLMs; F1-score for complex relational reasoning reaches 0.81; and the framework supports real-time interactive material discovery with verifiable knowledge provenance.

Technology Category

• Data Mining & Knowledge Management: Linked Open Data, Knowledge Graphs & KB Completion
• Machine Learning: Large Multimodal Models (LMMs)
• Knowledge Representation and Reasoning: Knowledge Representation Languages

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Vertical and domain-specific search

Abstract

Large language models (LLMs), such as ChatGPT, have demonstrated impressive performance in the text generation task, showing the ability to understand and respond to complex instructions. However, the performance of naive LLMs in speciffc domains is limited due to the scarcity of domain-speciffc corpora and specialized training. Moreover, training a specialized large-scale model necessitates signiffcant hardware resources, which restricts researchers from leveraging such models to drive advances. Hence, it is crucial to further improve and optimize LLMs to meet speciffc domain demands and enhance their scalability. Based on the condensed matter data center, we establish a material knowledge graph (MaterialsKG) and integrate it with literature. Using large language models and prompt learning, we develop a specialized dialogue system for topological materials called TopoChat. Compared to naive LLMs, TopoChat exhibits superior performance in structural and property querying, material recommendation, and complex relational reasoning. This system enables efffcient and precise retrieval of information and facilitates knowledge interaction, thereby encouraging the advancement on the ffeld of condensed matter materials.

Problem

• Large Language Models
• Domain-specific Knowledge
• Resource Efficiency

Innovation

• MaterialsKG Integration
• Domain-specific Performance Enhancement
• TopoChat Dialog System