Graph-R1: Towards Agentic GraphRAG Framework via End-to-End RL

20 hours ago 1

[Submitted on 29 Jul 2025]

View PDF HTML (experimental)

Abstract:Retrieval-Augmented Generation (RAG) mitigates hallucination in LLMs by incorporating external knowledge, but relies on chunk-based retrieval that lacks structural semantics. GraphRAG methods improve RAG by modeling knowledge as entity-relation graphs, but still face challenges in high construction cost, fixed one-time retrieval, and reliance on long-context reasoning and prompt design. To address these challenges, we propose Graph-R1, an agentic GraphRAG framework via end-to-end reinforcement learning (RL). It introduces lightweight knowledge hypergraph construction, models retrieval as a multi-turn agent-environment interaction, and optimizes the agent process via an end-to-end reward mechanism. Experiments on standard RAG datasets show that Graph-R1 outperforms traditional GraphRAG and RL-enhanced RAG methods in reasoning accuracy, retrieval efficiency, and generation quality.

Submission history

From: Haoran Luo [view email]
[v1] Tue, 29 Jul 2025 15:01:26 UTC (3,089 KB)

Read Entire Article