<ul data-eligibleForWebStory="true"><li>Teaching large language models (LLMs) to use tools effectively is crucial for improving their problem-solving abilities and expanding their applications.</li><li>Previous methods of generating instruction data for LLMs lacked in quality, as they relied on the LLMs themselves.</li><li>A new method proposed in this paper uses knowledge graphs to create high-quality instruction data for LLMs by extracting query pathways and translating relationships into actionable tools.</li><li>Experiments show that fine-tuning LLMs on synthetic data generated through knowledge graphs can significantly enhance their tool utilization and overall capabilities.</li></ul>

Enhancing LLM Tool Use with High-quality Instruction Data from Knowledge Graph

Discover more