KG-RAG: Bridging the Gap Between Knowledge and Creativity
This addresses the challenge of ensuring factual accuracy while maintaining creativity in intelligent agent systems, though it appears incremental as it builds on existing RAG and KGQA approaches.
The paper tackles the problem of factual inaccuracies and hallucinations in Large Language Model Agents when handling knowledge-intensive tasks by introducing KG-RAG, a pipeline that integrates Knowledge Graphs with LLMs. Preliminary experiments on ComplexWebQuestions show notable reductions in hallucinated content.
Ensuring factual accuracy while maintaining the creative capabilities of Large Language Model Agents (LMAs) poses significant challenges in the development of intelligent agent systems. LMAs face prevalent issues such as information hallucinations, catastrophic forgetting, and limitations in processing long contexts when dealing with knowledge-intensive tasks. This paper introduces a KG-RAG (Knowledge Graph-Retrieval Augmented Generation) pipeline, a novel framework designed to enhance the knowledge capabilities of LMAs by integrating structured Knowledge Graphs (KGs) with the functionalities of LLMs, thereby significantly reducing the reliance on the latent knowledge of LLMs. The KG-RAG pipeline constructs a KG from unstructured text and then performs information retrieval over the newly created graph to perform KGQA (Knowledge Graph Question Answering). The retrieval methodology leverages a novel algorithm called Chain of Explorations (CoE) which benefits from LLMs reasoning to explore nodes and relationships within the KG sequentially. Preliminary experiments on the ComplexWebQuestions dataset demonstrate notable improvements in the reduction of hallucinated content and suggest a promising path toward developing intelligent systems adept at handling knowledge-intensive tasks.