Graph-R1: Incentivizing the Zero-Shot Graph Learning Capability in LLMs via Explicit Reasoning

TL;DR

Graph-R1 leverages explicit reasoning with Large Reasoning Models to improve zero-shot graph task performance, offering a novel, interpretable approach that outperforms existing methods.

Contribution

Introduces Graph-R1, a reinforcement learning framework that reformulates graph tasks as textual reasoning problems for LRMs, with new datasets and templates for zero-shot graph learning.

Findings

Graph-R1 outperforms state-of-the-art baselines in zero-shot tasks.

Provides interpretable reasoning traces for graph tasks.

Demonstrates the effectiveness of explicit reasoning in graph learning.

Abstract

Generalizing to unseen graph tasks without task-pecific supervision remains challenging. Graph Neural Networks (GNNs) are limited by fixed label spaces, while Large Language Models (LLMs) lack structural inductive biases. Recent advances in Large Reasoning Models (LRMs) provide a zero-shot alternative via explicit, long chain-of-thought reasoning. Inspired by this, we propose a GNN-free approach that reformulates graph tasks--node classification, link prediction, and graph classification--as textual reasoning problems solved by LRMs. We introduce the first datasets with detailed reasoning traces for these tasks and develop Graph-R1, a reinforcement learning framework that leverages task-specific rethink templates to guide reasoning over linearized graphs. Experiments demonstrate that Graph-R1 outperforms state-of-the-art baselines in zero-shot settings, producing interpretable and effective predictions. Our work highlights the promise of explicit reasoning for graph learning and provides new resources for future research.