GraphReAct: Reasoning and Acting for Multi-step Graph Inference

TL;DR

GraphReAct introduces a step-by-step reasoning framework for graph data, combining retrieval and refinement actions to improve multi-step inference and outperform existing methods.

Contribution

It proposes a novel reasoning-acting framework with graph-specific actions for dynamic, multi-step graph inference, addressing the challenge of structured data reasoning.

Findings

Outperforms state-of-the-art methods on six benchmark datasets.

Effectively combines topological and semantic retrieval for graph reasoning.

Enables progressive context expansion and compression during inference.

Abstract

Reasoning-acting frameworks enhance large language models (LLMs) by interleaving reasoning with actions for dynamic information acquisition. However, extending this paradigm to graph learning remains underexplored. Graph data is inherently structured, with information distributed across nodes and edges and encoded through both topology and latent representations. As a result, effective reasoning over graphs requires not only retrieving informative evidence from the graph, but also progressively refining the accumulated context during multi-step inference. In this work, we propose GraphReAct, a graph reasoning-acting framework that enables step-by-step inference over graph-structured data. Specifically, we design a graph-based action space with two complementary retrieval actions: topological retrieval, which captures local structural dependencies, and semantic retrieval, which accesses non-local but relevant evidence in the representation space. These actions dynamically expand the reasoning context. To further support multi-step reasoning, we introduce another type of action, context refinement, which distills and reorganizes accumulated information into a compact representation. By interleaving reasoning with both retrieval and refinement actions, our framework enables a progressive transition from context expansion to compression. Extensive experiments on six benchmark datasets demonstrate that GraphReAct consistently outperforms state-of-the-art methods, validating the effectiveness of reasoning-acting for graph learning.