Graph-R1: Towards Agentic GraphRAG Framework via End-to-end Reinforcement Learning

Haoran Luo, Haihong E, Guanting Chen, Qika Lin, Yikai Guo +6 more
7/29/2025
cs.CL

No official code found. Best implementation: LHRLAB/Graph-R1 (582 stars, MIT, last commit 2026-04-30).

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.

View on arXivView PDF

Code Implementations(1)

LHRLAB/Graph-R1Community69%
MIT

[ICML 2026] Official resources of "Graph-R1: Towards Agentic GraphRAG Framework via End-to-end Reinforcement Learning".

58275Shell, PythonMar 11, 20255 months agoMIT
chain-of-thoughtgraphraghypergraphreinforcement-learning

Similar Papers

GraphRAG-R1: Graph Retrieval-Augmented Generation with Process-Constrained Reinforcement Learning

93% match2025

Efficient and Transferable Agentic Knowledge Graph RAG via Reinforcement Learning

90% match1 repo402025

ProGraph-R1: Progress-aware Reinforcement Learning for Graph Retrieval Augmented Generation

90% match2026

RouteRAG: Efficient Retrieval-Augmented Generation from Text and Graph via Reinforcement Learning

90% match2025

AutoGraph-R1: End-to-End Reinforcement Learning for Knowledge Graph Construction

88% match2025

Weak-to-Strong GraphRAG: Aligning Weak Retrievers with Large Language Models for Graph-based Retrieval Augmented Generation

87% match2025

Cite this paper

@article{luo2025graphr,
  title  = {Graph-R1: Towards Agentic GraphRAG Framework via End-to-end Reinforcement Learning},
  author = {Haoran Luo and Haihong E and Guanting Chen and Qika Lin and Yikai Guo and Fangzhi Xu and Zemin Kuang and Meina Song and Xiaobao Wu and Yifan Zhu and Luu Anh Tuan},
  year   = {2025},
  eprint = {2507.21892},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2507.21892v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2507.21892.svg)](https://codeofpaper.com/paper/2507.21892)

Discussion