Query-Aware Flow Diffusion for Graph-Based RAG with Retrieval Guarantees
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866909054912167936 |
|---|---|
| author | Zhou, Zhuoping Tarzanagh, Davoud Ataee Didari, Sima Hu, Wenjun Gutow, Baruch Verkholyak, Oxana Faraki, Masoud Hao, Heng Moon, Hankyu Min, Seungjai |
| author_facet | Zhou, Zhuoping Tarzanagh, Davoud Ataee Didari, Sima Hu, Wenjun Gutow, Baruch Verkholyak, Oxana Faraki, Masoud Hao, Heng Moon, Hankyu Min, Seungjai |
| contents | Graph-based Retrieval-Augmented Generation (RAG) systems leverage interconnected knowledge structures to capture complex relationships that flat retrieval struggles with, enabling multi-hop reasoning. Yet most existing graph-based methods suffer from (i) heuristic designs lacking theoretical guarantees for subgraph quality or relevance and/or (ii) the use of static exploration strategies that ignore the query's holistic meaning, retrieving neighborhoods or communities regardless of intent. We propose Query-Aware Flow Diffusion RAG (QAFD-RAG), a training-free framework that dynamically adapts graph traversal to each query's holistic semantics. The central innovation is query-aware traversal: during graph exploration, edges are dynamically weighted by how well their endpoints align with the query's embedding, guiding flow along semantically relevant paths while avoiding structurally connected but irrelevant regions. These query-specific reasoning subgraphs enable the first statistical guarantees for query-aware graph retrieval, showing that QAFD-RAG recovers relevant subgraphs with high probability under mild signal-to-noise conditions. The algorithm converges exponentially fast, with complexity scaling with the retrieved subgraph size rather than the full graph. Experiments on question answering and text-to-SQL tasks demonstrate consistent improvements over state-of-the-art graph-based RAG methods. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_18775 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Query-Aware Flow Diffusion for Graph-Based RAG with Retrieval Guarantees Zhou, Zhuoping Tarzanagh, Davoud Ataee Didari, Sima Hu, Wenjun Gutow, Baruch Verkholyak, Oxana Faraki, Masoud Hao, Heng Moon, Hankyu Min, Seungjai Information Retrieval Artificial Intelligence 68T50, 68T05, 68R10, 68Q25 H.3.3; I.2.7; I.2.6; G.2.2 Graph-based Retrieval-Augmented Generation (RAG) systems leverage interconnected knowledge structures to capture complex relationships that flat retrieval struggles with, enabling multi-hop reasoning. Yet most existing graph-based methods suffer from (i) heuristic designs lacking theoretical guarantees for subgraph quality or relevance and/or (ii) the use of static exploration strategies that ignore the query's holistic meaning, retrieving neighborhoods or communities regardless of intent. We propose Query-Aware Flow Diffusion RAG (QAFD-RAG), a training-free framework that dynamically adapts graph traversal to each query's holistic semantics. The central innovation is query-aware traversal: during graph exploration, edges are dynamically weighted by how well their endpoints align with the query's embedding, guiding flow along semantically relevant paths while avoiding structurally connected but irrelevant regions. These query-specific reasoning subgraphs enable the first statistical guarantees for query-aware graph retrieval, showing that QAFD-RAG recovers relevant subgraphs with high probability under mild signal-to-noise conditions. The algorithm converges exponentially fast, with complexity scaling with the retrieved subgraph size rather than the full graph. Experiments on question answering and text-to-SQL tasks demonstrate consistent improvements over state-of-the-art graph-based RAG methods. |
| title | Query-Aware Flow Diffusion for Graph-Based RAG with Retrieval Guarantees |
| topic | Information Retrieval Artificial Intelligence 68T50, 68T05, 68R10, 68Q25 H.3.3; I.2.7; I.2.6; G.2.2 |
| url | https://arxiv.org/abs/2605.18775 |