LLM Inference Enhanced by External Knowledge: A Survey

TL;DR

This survey reviews methods of integrating external structured knowledge, like tables and knowledge graphs, into large language models to improve reasoning, reduce hallucinations, and enhance trustworthiness and scalability.

Contribution

It provides a comprehensive taxonomy of external knowledge types and integration strategies, focusing on structured data and analyzing their trade-offs.

Findings

Structured knowledge integration improves reasoning accuracy.

Trade-offs exist between interpretability, scalability, and performance.

Comparative analysis guides future development of trustworthy LLMs.

Abstract

Recent advancements in large language models (LLMs) have enhanced natural-language reasoning. However, their limited parametric memory and susceptibility to hallucination present persistent challenges for tasks requiring accurate, context-based inference. To overcome these limitations, an increasing number of studies have proposed leveraging external knowledge to enhance LLMs. This study offers a systematic exploration of strategies for using external knowledge to enhance LLMs, beginning with a taxonomy that categorizes external knowledge into unstructured and structured data. We then focus on structured knowledge, presenting distinct taxonomies for tables and knowledge graphs (KGs), detailing their integration paradigms with LLMs, and reviewing representative methods. Our comparative analysis further highlights the trade-offs among interpretability, scalability, and performance, providing insights for developing trustworthy and generalizable knowledge-enhanced LLMs.