SAGE通过构建分块图并动态扩展，实现跨文本、表格和图谱的多跳证据检索。
SAGE: Structure Aware Graph Expansion for Retrieval of Heterogeneous Data
- 离线构建分块级图，基于元数据相似性与百分位剪枝
- 在线检索时先找种子段落，再扩展邻居并融合稠密/稀疏检索
- 适用于异构数据检索，尤其适合需跨模态推理的任务
在异构语料上进行检索增强型问答，需要跨越文本、表格和图节点的连贯证据。尽管实体级知识图谱支持结构化访问，但其构建和维护成本高，查询时遍历效率低。相比之下，标准检索-阅读流水线对独立切分的文本进行扁平相似度搜索，忽略了跨模态的多跳证据链。我们提出SAGE（结构感知图扩展）框架：(i) 离线使用元数据驱动的相似性构建分块级图，并采用百分位剪枝；(ii) 在线检索时，先通过基础检索器获取k个种子段落，扩展第一跳邻居，再结合稠密+稀疏检索过滤邻居，选出k'个新增段落。初始检索器采用混合稠密+稀疏检索处理隐式跨模态语料，或使用SPARK（面向知识图谱检索的结构感知规划代理）处理显式模式图。在OTT-QA和STaRK数据集上，SAGE相比基线分别提升检索召回率5.7和8.5个百分点。
原文摘要 · Abstract (English)
Retrieval-augmented question answering over heterogeneous corpora requires connected evidence across text, tables, and graph nodes. While entity-level knowledge graphs support structured access, they are costly to construct and maintain, and inefficient to traverse at query time. In contrast, standard retriever-reader pipelines use flat similarity search over independently chunked text, missing multi-hop evidence chains across modalities. We propose SAGE (Structure Aware Graph Expansion) framework that (i) constructs a chunk-level graph offline using metadata-driven similarities with percentile-based pruning, and (ii) performs online retrieval by running an initial baseline retriever to obtain k seed chunks, expanding first-hop neighbors, and then filtering the neighbors using dense+sparse retrieval, selecting k' additional chunks. We instantiate the initial retriever using hybrid dense+sparse retrieval for implicit cross-modal corpora and SPARK (Structure Aware Planning Agent for Retrieval over Knowledge Graphs) an agentic retriever for explicit schema graphs. On OTT-QA and STaRK, SAGE improves retrieval recall by 5.7 and 8.5 points over baselines.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。