analiza aiqbench
This paper introduces HyGRAG, a hierarchical graph framework that fuses entity and chunk representations to overcome the limitations of traditional similarity-based retrieval in RAG systems. By enabling context-aware searches across multiple abstraction levels with efficient dynamic updates, it significantly boosts multi-hop reasoning accuracy without sacrificing performance. The work represents a meaningful advancement in integrating structured knowledge graphs directly into large language model workflows for deeper information synthesis.
A Unified Framework for Context-Aware and Relation-Aware Graph Retrieval-Augmented Generation
arXiv:2606.18075v1 Announce Type: new Abstract: Retrieval-Augmented Generation (RAG) has emerged as a paradigm for enhancing large language models (LLMs) with external knowledge, yet existing graph-based methods face a fundamental limitation: entity-centric and chunk-centric approaches operate on representations anchored to original text without true knowledge fusion. While entity-centric methods connect logically related content and chunk-centric methods preserve context, both retrieve information separately through similarity search, missing emergent understanding from their synthesis. In this paper, we propose HyGRAG, a hierarchical graph RAG framework that transcends source documents by addressing three core challenges: constructing summaries that genuinely integrate contextual and relational information, leveraging these synthesized representations to access emergent knowledge during retrieval, and efficiently updating hierarchical structures for dynamic corpora. Specifically, we design hierarchical index structures over hybrid graphs with both chunk and entity nodes, then iteratively cluster them and generate LLM-based summaries. Then, we design context and relation-aware retrieval that searches across all abstraction levels while expanding through community membership. Moreover, we enable dynamic knowledge update through attachment-based algorithms with only local re-summarization. Experimental results show that HyGRAG improves the average accuracy of multi-hop reasoning tasks by 9.7%, while maintaining reasonable efficiency.
Abstract:Retrieval-Augmented Generation (RAG) has emerged as a paradigm for enhancing large language models (LLMs) with external knowledge, yet existing graph-based methods face a fundamental limitation: entity-centric and chunk-centric approaches operate on representations anchored to original text without true knowledge fusion. While entity-centric methods connect logically related content and chunk-centric methods preserve context, both retrieve information separately through similarity search, missing emergent understanding from their synthesis. In this paper, we propose HyGRAG, a hierarchical graph RAG framework that transcends source documents by addressing three core challenges: constructing summaries that genuinely integrate contextual and relational information, leveraging these synthesized representations to access emergent knowledge during retrieval, and efficiently updating hierarchical structures for dynamic corpora. Specifically, we design hierarchical index structures over hybrid graphs with both chunk and entity nodes, then iteratively cluster them and generate LLM-based summaries. Then, we design context and relation-aware retrieval that searches across all abstraction levels while expanding through community membership. Moreover, we enable dynamic knowledge update through attachment-based algorithms with only local re-summarization. Experimental results show that HyGRAG improves the average accuracy of multi-hop reasoning tasks by 9.7%, while maintaining reasonable efficiency.
Artificial Intelligence (cs.AI)
arXiv:2606.18075 [cs.AI]
(or arXiv:2606.18075v1 [cs.AI] for this version)
https://doi.org/10.48550/arXiv.2606.18075 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
https://doi.org/10.1145/3774904.3792720 Focus to learn more DOI(s) linking to related resources
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 289 cuvinte
Subiecte conexe
LlamaRec-LKG-RAG: A Single-Pass, Learnable Knowledge Graph-RAG Framework for LLM-Based Ranking
arXiv · 18 aug. 2026
Sheet As Token: A Graph-Enhanced Representation for Multi-Sheet Spreadsheet Understanding
arXiv · 29 iul. 2026
Agent-UCT: Upper Confidence Bounds Applied to Trees for Agentic Workflow Optimization with Cost-Awareness
arXiv · 28 iul. 2026
Integrating Factual and Normative Industrial Knowledge via Constraint-Aware Graph Attention for Process Plan Recommendation
arXiv · 28 iul. 2026