通过图传播机制提升长文档中证据页的检索精度
EviProp: Seeded Relevance Diffusion on Chunk-Page Graphs for Long Multimodal Document Retrieval
- 构建图文块-页图结构，用种子信息扩散增强相关页
- 在两个数据集上相比独立检索提升显著，最高增益达12.3%
- 适合需要精准定位长文档中细粒度证据的研究者
从视觉丰富的长文档中检索证据页是文档问答的关键挑战。现有页面级视觉检索器采用独立匹配范式：每个页面独立根据查询-页面相似性打分。该范式会低估信号集中在细粒度块或依赖文档内部关联的证据页。我们提出EviProp，一种通过种子相关性传播恢复此类页面的检索方法。EviProp将每个文档建模为具有层级、顺序和相似性链接的多模态块-页图。给定查询后，结合密集视觉页面先验与稀疏块种子，运行个性化PageRank在图上扩散相关性。在MMLongBench-Doc和LongDocURL上的实验表明，相比独立视觉检索和文本-视觉融合基线，证据页检索性能持续提升。下游问答结果进一步显示，检索改进带来更高答案准确率，且在线检索开销可忽略。代码已开源：https://github.com/Flyecnu/EviProp。
原文摘要 · Abstract (English)
Retrieving evidence pages from visually rich long documents is a key challenge in document question answering. Existing page-level visual retrievers operate under an independent matching paradigm: each page is scored in isolation based on query-page similarity. This paradigm can under-rank evidence pages whose signals are localized in fine-grained chunks or depend on document-internal associations. We propose EviProp, a retrieval method that recovers such pages via seeded relevance diffusion. EviProp models each document as a multimodal Chunk-Page graph with hierarchical, sequential, and similarity links. Given a query, it combines dense visual page priors with sparse chunk seeds, then runs Personalized PageRank to diffuse relevance over the graph. Experiments on MMLongBench-Doc and LongDocURL show consistent gains in evidence-page retrieval over independent visual retrieval and text-visual fusion baselines. Downstream QA results further show that improved retrieval translates into better answer accuracy, with negligible online retrieval overhead. Our code is released at https://github.com/Flyecnu/EviProp.
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。