{"ID":2900916,"CreatedAt":"2026-06-01T05:51:17.9442275Z","UpdatedAt":"2026-06-01T06:23:29.641557848Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2605.31010","arxiv_id":"2605.31010","title":"MoG: Mixture of Experts for Graph-based Retrieval-Augmented Generation","abstract":"Retrieval-augmented generation is intensively studied to ground large language models on external evidence. However, retrieving from a unified knowledge base could inevitably introduce irrelevant information that may mislead generation for complex reasoning. Inspired by the conditional computation of mixture of experts (MoE), where a router sparsely selects specialized experts alongside shared ones for each input, we propose \\textbf{M}ixture \\textbf{o}f experts for \\textbf{G}raph-based Retrieval-Augmented Generation, i.e., \\textbf{MoG}. It organizes knowledge into two core components: (i) diverse, always-accessible hub graphs that encode semantically and structurally central knowledge and provide contextual clues for expert activation, and (ii) sparsely activated expert graphs that contain domain-specific evidence. MoG first accesses hub graphs to identify general evidence and derive contextual clues. Then, a topology-aware router dynamically activates a limited set of expert graphs conditioned on the query, thereby confining retrieval to a focused evidence subspace. Extensive experiments on challenging benchmarks show that MoG consistently outperforms strong baselines, with over 20\\% relative improvement on MuSiQue. Our code is available in https://github.com/DEEP-PolyU/MoG.","short_abstract":"Retrieval-augmented generation is intensively studied to ground large language models on external evidence. However, retrieving from a unified knowledge base could inevitably introduce irrelevant information that may mislead generation for complex reasoning. Inspired by the conditional computation of mixture of experts...","url_abs":"https://arxiv.org/abs/2605.31010","url_pdf":"https://arxiv.org/pdf/2605.31010v1","authors":"[\"Zheng Yuan\",\"Chuang Zhou\",\"Linhao Luo\",\"Siyu An\",\"Di Yin\",\"Xing Sun\",\"Xiao Huang\"]","published":"2026-05-29T08:43:12Z","proceeding":"cs.CL","tasks":"[\"cs.CL\"]","methods":"[\"RAG\",\"Mixture of Experts\",\"Language Model\",\"Generative Adversarial Network\"]","has_code":false,"code_links":[{"ID":612543,"CreatedAt":"2026-06-01T05:51:17.9442275Z","UpdatedAt":"2026-06-01T05:51:17.9442275Z","DeletedAt":null,"paper_id":2900916,"paper_url":"https://arxiv.org/abs/2605.31010","paper_title":"MoG: Mixture of Experts for Graph-based Retrieval-Augmented Generation","repo_url":"https://github.com/DEEP-PolyU/MoG","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}