Scientific Knowledge Graph and Ontology Generation using Open Large Language Models

Key Points

• The research aims to explore how knowledge graphs can improve the factual accuracy and reasoning skills of large language models.
• Analysis of existing knowledge graph frameworks
• Integration of large language models with structured information
• Evaluation of reasoning capabilities through practical tasks
• Knowledge graphs significantly improved the factual accuracy of large language models.
• The reasoning capabilities of models using integrated knowledge graphs were notably enhanced.
• Models showed increased performance in tasks requiring factual and contextual understanding.