混合向量检索兼顾效率与精度，适合复杂文档搜索。
Hybrid-Vector Retrieval for Visually Rich Documents: Combining Single-Vector Efficiency and Multi-Vector Accuracy
- 两阶段设计：先用单向量快速筛选，再用多向量精排。
- 在四组数据上达多向量99.87%召回率，计算量减少99.82%。
- 适用于法律、科研等需高精度的长文档检索场景。
视觉丰富文档的检索对法律发现、科学搜索和企业知识管理至关重要。现有方法分为两类：单向量检索效率高但粗略，多向量检索准确但计算成本高。为解决这一权衡，我们提出HEAVEN，一种即插即用的两阶段混合向量框架。第一阶段，基于视觉摘要页（VS-Pages）的单向量方法高效检索候选页面；第二阶段，通过按语言重要性过滤查询词，以减少冗余计算，使用多向量方法重排序候选结果。为在真实条件下评估检索系统，我们还引入ViMDoc基准，用于视觉丰富、多文档及长文档检索。在四个基准上，HEAVEN平均达到多向量模型99.87%的Recall@1性能，同时每查询计算量降低99.82%，实现高效与精准的平衡。代码与数据集已公开于：https://github.com/juyeonnn/HEAVEN
原文摘要 · Abstract (English)
Retrieval over visually rich documents is essential for tasks such as legal discovery, scientific search, and enterprise knowledge management. Existing approaches fall into two paradigms: single-vector retrieval, which is efficient but coarse, and multi-vector retrieval, which is accurate but computationally expensive. To address this trade-off, we propose HEAVEN, a plug-and-play two-stage hybrid-vector framework. In the first stage, HEAVEN efficiently retrieves candidate pages using a single-vector method over Visually-Summarized Pages (VS-Pages), which assemble representative visual layouts from multiple pages. In the second stage, it reranks candidates with a multi-vector method while filtering query tokens by linguistic importance to reduce redundant computations. To evaluate retrieval systems under realistic conditions, we also introduce ViMDoc, a benchmark for visually rich, multi-document, and long-document retrieval. Across four benchmarks, HEAVEN attains 99.87% of the Recall@1 performance of multi-vector models on average while reducing per-query computation by 99.82%, achieving efficiency and accuracy. Our code and datasets are available at: https://github.com/juyeonnn/HEAVEN
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。