发表机构
South China University of Technology; Foshan University（华南理工大学; 佛山大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
提出G$^3$RAG，利用几何增益分数构建零令牌文档图，无需LLM实体抽取，在多跳RAG基准上超越现有图方法，提升F1并降低成本。
AI 中文摘要
多跳检索增强生成（RAG）需要证据既与查询保持相关性，又引入足够的新颖性以弥合语义鸿沟。密集检索倾向于聚焦于语义相似的文档，而基于图的方法通常依赖昂贵的大语言模型（LLM）实体抽取，并可能通过噪声连接进行传播。我们提出了几何增益图RAG（G$^3$RAG），一个仅基于文档的框架，其离线图构建不使用任何LLM调用或生成令牌。G$^3$RAG为每条边分配一个几何增益分数，即$\cos\theta \cdot \sin\theta$，该分数联合捕捉文档表示之间的方向一致性和正交性。一个密度感知的拓扑惩罚项抑制高度连接的枢纽节点，而单步受控扩散从过滤后的查询种子向互补证据扩展。我们在MusiQue、2WikiMultiHopQA和HotpotQA上使用Nv-embed-v2和Qwen3-8B-embed评估了G$^3$RAG。在两种嵌入设置中，G$^3$RAG在评估的基于图的基线中获得了最佳的平均F1分数和答案-文档命中率，平均性能提升高达4.26个F1点，在MusiQue上提升5.76个点。它还消除了基于实体的图方法所产生的图构建令牌成本。这些结果表明，几何结构可以在不依赖基于LLM的图构建的情况下支持高效的多跳证据发现。代码可在以下https URL获取。
英文摘要
Multi-hop retrieval-augmented generation (RAG) requires evidence that remains relevant to a query while introducing enough novelty to bridge semantic gaps. Dense retrieval tends to concentrate on semantically similar documents, whereas graph-based alternatives often depend on costly Large Language Model (LLM) entity extraction and may propagate through noisy connections. We introduce Geometric Gain Graph RAG (G$^3$RAG), a document-only framework whose offline graph construction uses no LLM calls or generated tokens. G$^3$RAG assigns each edge a geometric gain score, $\cosθ\cdot \sinθ$, that jointly captures directional consistency and orthogonality between document representations. A density-aware topological penalty suppresses highly connected hubs, while single-step controlled diffusion expands from filtered query seeds toward complementary evidence. We evaluate G$^3$RAG on MusiQue, 2WikiMultiHopQA, and HotpotQA using Nv-embed-v2 and Qwen3-8B-embed. G$^3$RAG obtains the best average F1 and answer-document hit rate among the evaluated graph-based baselines in both embedding settings, with gains of up to 4.26 F1 points in average performance and 5.76 points on MusiQue. It also removes the graph-construction token cost incurred by entity-based graph methods. These results show that geometric structure can support efficient multi-hop evidence discovery without LLM-based graph construction. Code is available at https://anonymous.4open.science/r/G3RAG-99D9/
