KGiRAG: An Iterative GraphRAG Approach for Responding Sensemaking Queries

TL;DR

This paper introduces KGiRAG, an iterative graph-based retrieval-augmented generation method that refines responses through feedback to improve answer quality for complex queries.

Contribution

The paper presents a novel iterative GraphRAG architecture that enhances LLM response quality by leveraging feedback-driven refinement for complex query answering.

Findings

Iterative GraphRAG produces higher semantic quality responses.

The approach improves relevance over single-shot methods.

Evaluation on HotPotQA shows significant performance gains.

Abstract

Recent literature highlights the potential of graph-based approaches within large language model (LLM) retrieval-augmented generation (RAG) pipelines for answering queries of varying complexity, particularly those that fall outside the LLM's prior knowledge. However, LLMs are prone to hallucination and often face technical limitations in handling contexts large enough to ground complex queries effectively. To address these challenges, we propose a novel iterative, feedback-driven GraphRAG architecture that leverages response quality assessment to iteratively refine outputs until a sound, well-grounded response is produced. Evaluating our approach with queries from the HotPotQA dataset, we demonstrate that this iterative RAG strategy yields responses with higher semantic quality and improved relevance compared to a single-shot baseline.