Knowledge Tree
knowledge-graph-enhanced LLM
Also known as: knowledge graph-augmented LLM, knowledge-graph-enhanced LLM system, knowledge-graph-enhanced LLMs, knowledge-graph-augmented LLM
Facts (23)
Sources
LLM-Powered Knowledge Graphs for Enterprise Intelligence and ... arxiv.org Mar 11, 2025 18 facts
measurementThe knowledge-graph-enhanced LLM system achieved 92% accuracy in entity extraction and 89% accuracy in relationship extraction, with contextual enrichment improving task alignment by 15%.
measurementThe knowledge-graph-enhanced LLM system achieved 86% accuracy when comparing system-generated analytics to manual calculations.
procedureTo quantitatively assess the performance of the knowledge-graph-enhanced LLM system, the researchers used NDCG (Normalized Discounted Cumulative Gain) for rank-based tasks, Mean Reciprocal Rank (MRR) for precision of top-ranked results, and additional indicators including user satisfaction rates, precision, and recall.
claimThe knowledge-graph-enhanced LLM system facilitates data-driven decision-making by leveraging graph analytics and LLMs to translate natural language queries into graph traversal and analytics operations, allowing for the retrieval and segmentation of relevant statistics.
measurementThe knowledge-graph-enhanced LLM system is currently being tested for more than five use cases, despite the documentation detailing only three key applications.
procedureThe knowledge-graph-enhanced LLM system answers analytics queries by retrieving statistics from the knowledge graph, refining the data via the LLM, and generating actionable insights.
measurementThe knowledge-graph-enhanced LLM system achieved an MPR (Mean Position of Relevant expert) of 0.83 for expert recommendations.
claimA knowledge-graph-enhanced LLM system improves employee productivity and task prioritization by traversing a knowledge graph to provide daily or weekly task recommendations and displaying relevant contextual materials or conversations.
measurementOver a six-month period, the knowledge-graph-enhanced LLM system achieved a 78% user adoption rate across multiple departments and successfully addressed five of six targeted scenarios, including expertise discovery, task prioritization, and analytics.
accountA 6-month pilot study of the knowledge-graph-enhanced LLM system was conducted in two large companies within the finance and healthcare sectors to validate performance in real-world environments.
measurementThe knowledge-graph-enhanced LLM system achieved an NDCG@5 of 0.80 and an NDCG@3 of 0.63 for expert recommendation ranking quality.
measurementThe knowledge-graph-enhanced LLM system achieved an NDCG@5 of 0.72 and an NDCG@3 of 0.59 for task prioritization based on implicit user feedback.
measurementThe knowledge-graph-enhanced LLM system achieved a Precision at K of P@3=0.62 and P@5=0.83 for expert recommendations.
referenceThe Recommendations and Analytics layer in the knowledge-graph-enhanced LLM system combines knowledge graph data with LLM-based reasoning to provide actionable insights and analytics for enterprise needs.
procedureThe knowledge-graph-enhanced LLM system prioritizes tasks by analyzing importance, urgency, and dependencies, while tracking user actions like task completions to infer implicit relevance signals.
measurementThe knowledge-graph-enhanced LLM system achieved an 83% positive feedback rate regarding user satisfaction with analytical insights.
measurementThe knowledge-graph-enhanced LLM system achieved a Recall of 0.83 for including critical tasks in recommendations.
measurementThe knowledge-graph-enhanced LLM system achieved a Precision at K of P@3=0.57 and P@5=0.80 for task prioritization alignment with completed tasks.
Leveraging Knowledge Graphs and LLM Reasoning to Identify ... arxiv.org Jul 23, 2025 2 facts
claimThe industrial use case scenario for knowledge-graph-enhanced LLMs involves analyzing supplier discharge times, package waiting times, and equipment utilization rates within a specific operational window of 10:00 AM to 12:30 PM.
procedureThe proposed technique for knowledge-graph-enhanced LLMs avoids the brittleness of using a single, monolithic Cypher query by implementing a layer based on question decomposition and structured step-wise guidance generation. This agent breaks down each query into a sequence of analytical steps, where each step involves targeted Cypher query formulation, execution, and an immediate self-reflection phase to assess and refine the output before proceeding.
Combining Knowledge Graphs With LLMs | Complete Guide - Atlan atlan.com Jan 28, 2026 2 facts
claimMulti-hop reasoning in knowledge graph-augmented LLM architectures enables answering questions that require linking information across several related entities.
procedureThe architecture for knowledge graph-augmented LLM systems operates in three stages: first, query understanding translates natural language into graph concepts; second, graph traversal follows relationships to collect relevant subgraphs; third, context assembly combines retrieved graph data with the original query for LLM processing.
Applying Large Language Models in Knowledge Graph-based ... arxiv.org Jan 7, 2025 1 fact
referenceResearch into employing knowledge graphs to address modeling language semantics includes three areas: (1) Knowledge Graph-enhanced LLMs for improving LLM knowledge during pre-training and inference, (2) LLM-augmented Knowledge Graphs for tasks like graph construction or question answering, and (3) Synergized LLMs + Knowledge Graphs for bidirectional enhancement of both systems.