Empowering GraphRAG with Knowledge Filtering and Integration

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Guo, Kai, Shomer, Harry, Zeng, Shenglai, Han, Haoyu, Wang, Yu, Tang, Jiliang
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913742478901248
author Guo, Kai
Shomer, Harry
Zeng, Shenglai
Han, Haoyu
Wang, Yu
Tang, Jiliang
author_facet Guo, Kai
Shomer, Harry
Zeng, Shenglai
Han, Haoyu
Wang, Yu
Tang, Jiliang
contents In recent years, large language models (LLMs) have revolutionized the field of natural language processing. However, they often suffer from knowledge gaps and hallucinations. Graph retrieval-augmented generation (GraphRAG) enhances LLM reasoning by integrating structured knowledge from external graphs. However, we identify two key challenges that plague GraphRAG:(1) Retrieving noisy and irrelevant information can degrade performance and (2)Excessive reliance on external knowledge suppresses the model's intrinsic reasoning. To address these issues, we propose GraphRAG-FI (Filtering and Integration), consisting of GraphRAG-Filtering and GraphRAG-Integration. GraphRAG-Filtering employs a two-stage filtering mechanism to refine retrieved information. GraphRAG-Integration employs a logits-based selection strategy to balance external knowledge from GraphRAG with the LLM's intrinsic reasoning,reducing over-reliance on retrievals. Experiments on knowledge graph QA tasks demonstrate that GraphRAG-FI significantly improves reasoning performance across multiple backbone models, establishing a more reliable and effective GraphRAG framework.
format Preprint
id arxiv_https___arxiv_org_abs_2503_13804
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Empowering GraphRAG with Knowledge Filtering and Integration
Guo, Kai
Shomer, Harry
Zeng, Shenglai
Han, Haoyu
Wang, Yu
Tang, Jiliang
Artificial Intelligence
In recent years, large language models (LLMs) have revolutionized the field of natural language processing. However, they often suffer from knowledge gaps and hallucinations. Graph retrieval-augmented generation (GraphRAG) enhances LLM reasoning by integrating structured knowledge from external graphs. However, we identify two key challenges that plague GraphRAG:(1) Retrieving noisy and irrelevant information can degrade performance and (2)Excessive reliance on external knowledge suppresses the model's intrinsic reasoning. To address these issues, we propose GraphRAG-FI (Filtering and Integration), consisting of GraphRAG-Filtering and GraphRAG-Integration. GraphRAG-Filtering employs a two-stage filtering mechanism to refine retrieved information. GraphRAG-Integration employs a logits-based selection strategy to balance external knowledge from GraphRAG with the LLM's intrinsic reasoning,reducing over-reliance on retrievals. Experiments on knowledge graph QA tasks demonstrate that GraphRAG-FI significantly improves reasoning performance across multiple backbone models, establishing a more reliable and effective GraphRAG framework.
title Empowering GraphRAG with Knowledge Filtering and Integration
topic Artificial Intelligence
url https://arxiv.org/abs/2503.13804