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

ICLR 2026 Conference Submission13931 Authors

Keywords: Knowledge Graph, Large Language Model, GraphRAG, Reinforcement Learning

TL;DR: Graph-R1 presents an agentic GraphRAG framework with end-to-end reinforcement learning that outperforms previous GraphRAG and RAG methods in accuracy, retrieval efficiency, and generation quality.

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. Our code is available.

Primary Area: neurosymbolic & hybrid AI systems (physics-informed, logic & formal reasoning, etc.)

Submission Number: 13931