arXiv cs.AI· Federica Santi (Cognee Inc), Vasilije Markovi\'{c} (Cognee Inc), Daniel Scott (Cognee Inc), Lazar Obradovi\'{c} (Cognee Inc)·· 3 天前AI 评分31
无图检索：用图修改嵌入提升稠密向量检索召回
Graph-free Retrieval with Graph-Modified Embeddings
AI 导读
研究比较了五种离线图嵌入变换及图修改嵌入与原始嵌入的搭配策略，无需查询时访问图。在合成职业网络基准上，直接替换原始嵌入会降低证据召回，而多阶段策略用图修改嵌入引导搜索、保留原始嵌入维持语义相关性，在需要关联事实或候选比较的问题上召回率相对纯向量检索最高提升 25.9%，并以少 40% 的返回节点数达到基线召回。检索表现更取决于图修改嵌入的使用方式，而非生成方式。
正文 · 原文
Abstract:Dense vector retrieval can miss evidence connected through intermediate entities, while direct graph retrieval requires access to the graph at query time. We compare five offline graph-based embedding transformations and strategies for using the resulting graph-modified embeddings alongside the originals, without query-time graph access. On a synthetic professional-network benchmark with complete supporting-evidence sets, direct replacement of the original embeddings reduces evidence recovery. Multi-stage strategies improve recovery by using graph-modified embeddings to guide the search while retaining the originals for semantic relevance. On questions requiring linked facts or comparisons among candidates, the strongest configuration improves recovery by up to 25.9% relative to plain vector retrieval. It also matches baseline recovery with 40% fewer returned nodes. Controls confirm that graph information contributes beyond multi-stage retrieval alone. Across the tested methods, retrieval performance depends more on how graph-modified embeddings are used than on how they are produced.
| Comments: | 28 pages, 10 figures, including appendices |
| Subjects: | Information Retrieval (cs.IR); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.18410 [cs.IR] |
| (or arXiv:2610.18410v1 [cs.IR] for this version) | |
| https://doi.org/10.48550/arXiv.2610.18410 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Lazar Obradovic [view email]
[v1]
Tue, 13 Oct 2026 13:51:46 UTC (38 KB)
来源：arXiv cs.AI · arxiv.org