Why Neighborhoods Matter: Traversal Context and Provenance in Agentic GraphRAG
For researchers and practitioners of retrieval-augmented generation, this work highlights a previously overlooked aspect of citation faithfulness in agentic systems, proposing a shift toward trajectory-level provenance evaluation.
The paper identifies that in Agentic GraphRAG, citation faithfulness depends not only on cited entities but also on the broader graph traversal context. Through ablation experiments, they show that removing cited evidence substantially changes answers and reduces accuracy, but accurate answers can also depend on uncited traversal context and graph structure.
Retrieval-Augmented Generation can improve factuality by grounding answers in external evidence, but Agentic GraphRAG complicates what it means for citations to be faithful. In these systems, an agent explores a knowledge graph before producing an answer and a small set of citations. We frame citation faithfulness as a trajectory-level problem: final citations should not only support the answer, but also account for the graph traversal, structure, and visited-but-uncited entities that may influence it. Through controlled ablation experiments, we compare the effects of isolating, removing, and masking cited and uncited graph entities. Our results show that cited evidence is often necessary, as removing it substantially changes answers and reduces accuracy. However, citations are not sufficient, because accurate answers can also depend on uncited traversal context and surrounding graph structure. These findings suggest that citation evaluation in Agentic GraphRAG should move beyond source support toward provenance over the broader retrieval trajectory.