{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,18]],"date-time":"2026-03-18T03:33:52Z","timestamp":1773804832549,"version":"3.50.1"},"reference-count":0,"publisher":"Association for the Advancement of Artificial Intelligence (AAAI)","issue":"37","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["AAAI"],"abstract":"<jats:p>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 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\u2019s low signal-to-noise ratio introduces distractor facts, presenting query-relevant but misleading knowledge that disrupts the reasoning process.\nTo address these challenges, we argue for a reason-and-construct paradigm and propose Relink, a framework that dynamically builds a query-specific evidence graph. To tackle incompleteness, 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.\nExtensive 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.<\/jats:p>","DOI":"10.1609\/aaai.v40i37.40382","type":"journal-article","created":{"date-parts":[[2026,3,18]],"date-time":"2026-03-18T02:48:26Z","timestamp":1773802106000},"page":"31202-31210","source":"Crossref","is-referenced-by-count":0,"title":["Relink: Constructing Query-Driven Evidence Graph On-the-Fly for GraphRAG"],"prefix":"10.1609","volume":"40","author":[{"given":"Manzong","family":"Huang","sequence":"first","affiliation":[]},{"given":"Chenyang","family":"Bu","sequence":"additional","affiliation":[]},{"given":"Yi","family":"He","sequence":"additional","affiliation":[]},{"given":"Xingrui","family":"Zhuo","sequence":"additional","affiliation":[]},{"given":"Xindong","family":"Wu","sequence":"additional","affiliation":[]}],"member":"9382","published-online":{"date-parts":[[2026,3,14]]},"container-title":["Proceedings of the AAAI Conference on Artificial Intelligence"],"original-title":[],"link":[{"URL":"https:\/\/ojs.aaai.org\/index.php\/AAAI\/article\/download\/40382\/44343","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/ojs.aaai.org\/index.php\/AAAI\/article\/download\/40382\/44343","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,3,18]],"date-time":"2026-03-18T02:48:26Z","timestamp":1773802106000},"score":1,"resource":{"primary":{"URL":"https:\/\/ojs.aaai.org\/index.php\/AAAI\/article\/view\/40382"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,3,14]]},"references-count":0,"journal-issue":{"issue":"37","published-online":{"date-parts":[[2026,3,17]]}},"URL":"https:\/\/doi.org\/10.1609\/aaai.v40i37.40382","relation":{},"ISSN":["2374-3468","2159-5399"],"issn-type":[{"value":"2374-3468","type":"electronic"},{"value":"2159-5399","type":"print"}],"subject":[],"published":{"date-parts":[[2026,3,14]]}}}