GraphSearch: An Agentic Deep Searching Workflow for Graph Retrieval-Augmented Generation

ACL ARR 2026 January Submission762 Authors

Keywords: Retrieval-Augmented Generation, GraphRAG, Large Language Model

Abstract: Graph Retrieval-Augmented Generation enhances factual reasoning in large language models by structurally modeling knowledge through graph-based representations. Existing GraphRAG approaches face two core limitations: shallow retrieval that fails to surface all critical evidence, and inefficient utilization of pre-constructed structural graph data. To address these challenges, we propose GraphSearch, a novel agentic deep searching workflow with dual-channel retrieval for GraphRAG. GraphSearch organizes the retrieval process into a modular framework comprising six modules, enabling multi-turn interactions and iterative reasoning. Furthermore, GraphSearch adopts a dual-channel retrieval strategy that issues semantic queries over chunk-based text data and relational queries over structural graph data, enabling comprehensive utilization of both modalities and their complementary strengths. Experimental results across six multi-hop RAG benchmarks demonstrate that GraphSearch consistently improves accuracy and generation quality over the traditional strategy, confirming GraphSearch as a promising direction for advancing agentic graph retrieval-augmented generation.

Paper Type: Long

Research Area: Retrieval-Augmented Language Models

Research Area Keywords: retrieval-augmented generation, LLM/AI agents

Contribution Types: NLP engineering experiment

Languages Studied: English

Submission Number: 762