English

Relink: Constructing Query-Driven Evidence Graph On-the-Fly for GraphRAG

Computation and Language 2026-01-13 v1 Artificial Intelligence

Abstract

Graph-based Retrieval-Augmented Generation (GraphRAG) mitigates hallucinations in Large Language Models (LLMs) by grounding them in structured knowledge. However, current GraphRAG methods are constrained by a prevailing \textit{build-then-reason} paradigm, which relies on a static, pre-constructed Knowledge Graph (KG). This paradigm faces two critical challenges. First, the KG's inherent incompleteness often breaks reasoning paths. Second, the graph's low signal-to-noise ratio introduces distractor facts, presenting query-relevant but misleading knowledge that disrupts the reasoning process. To address these challenges, we argue for a \textit{reason-and-construct} paradigm and propose Relink, a framework that dynamically builds a query-specific evidence graph. To tackle incompleteness, \textbf{Relink} instantiates required facts from a latent relation pool derived from the original text corpus, repairing broken paths on the fly. To handle misleading or distractor facts, Relink employs a unified, query-aware evaluation strategy that jointly considers candidates from both the KG and latent relations, selecting those most useful for answering the query rather than relying on their pre-existence. This empowers Relink to actively discard distractor facts and construct the most faithful and precise evidence path for each query. Extensive experiments on five Open-Domain Question Answering benchmarks show that Relink achieves significant average improvements of 5.4\% in EM and 5.2\% in F1 over leading GraphRAG baselines, demonstrating the superiority of our proposed framework.

Keywords

Cite

@article{arxiv.2601.07192,
  title  = {Relink: Constructing Query-Driven Evidence Graph On-the-Fly for GraphRAG},
  author = {Manzong Huang and Chenyang Bu and Yi He and Xingrui Zhuo and Xindong Wu},
  journal= {arXiv preprint arXiv:2601.07192},
  year   = {2026}
}

Comments

Accepted by AAAI 2026

R2 v1 2026-07-01T09:00:03.997Z