Position: How can Graphs Help Large Language Models?
For researchers and practitioners working with LLMs, this paper provides a structured overview of how graphs can address key limitations like hallucinations and reasoning, though it is a conceptual position rather than an empirical contribution.
This position paper explores how graphs can enhance large language models (LLMs) by reducing hallucinations through up-to-date knowledge, improving reasoning via graph-based prompting (e.g., Chain-of-Thought, Tree-of-Thought, Graph-of-Thought), and expanding applicability to structured data domains. It also outlines future directions such as sparse LLM architectures and brain-inspired memory systems.
With the rapid advancement of large language models (LLMs), classic graph learning tasks have greatly benefited from LLMs, including improved encoding of textual features, more efficient construction of graphs from text, and enhanced reasoning over knowledge graphs. In this paper, we ask a complementary question: How can graphs help LLMs? We address this question from three perspectives: 1) graphs provide an up-to-date knowledge source that helps reduce LLM hallucinations, 2) graph-based prompting techniques-such as Chain-of-Thought (CoT), Tree-of-Thought (ToT), and Graph-of-Thought (GoT)-enhance LLM reasoning capabilities, and 3) integrating graphs into LLMs improves their understanding of structured data, expanding their applicability to domains such as e-commerce, code, and relational databases (RDBs). We further outlook some future directions including designing sparse LLM architectures based on graphs and brain-inspired memory systems.