发表机构
United Nations University（联合国大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
研究RAG系统中检索问题，提出目录引导的页面检索新算法，通过实验发现该算法对答案质量有显著影响，与答案侧验证结合优于查询侧分解+验证，证明文档、查询、答案侧增强互补，且默认设置接近最优。
AI 中文摘要
RAG系统依赖分块，会破坏文档中的结构信息。现有的基于标题的检索需要对每个文档进行多次大语言模型调用并返回匹配部分内的子块。我们引入了目录（ToC）引导的页面检索，它无需大语言模型调用即可从视觉格式推断标题，将其作为并行索引嵌入并加载整页部分。在8个企业文档（5至195页）的1280个条件下，我们发现：（1）目录对答案质量有显著的主要影响（d = +0.41，p = 0.031），在完整性（+0.40）和有用性（+0.40）方面提升最大；（2）与答案侧验证相结合，它优于查询侧分解+验证（d = +0.32，p = 0.036）；（3）目录虽每个查询仅增加2.9页，但贡献了20%的引用；（4）对较长文档的提升更大（118页的文档提升高达+1.50），但8个文档的这一趋势未达显著水平；（5）480个条件的敏感性分析未发现显著的参数效应（所有p > 0.38），确认默认设置接近最优。该研究贡献既有方法层面（一种新的零大语言模型成本检索算法），也有实证层面：证明文档侧、查询侧和答案侧增强具有互补性的因子证据，这是此前未研究过的三向交互作用。
英文摘要
RAG systems rely on chunking, which destroys structural information in documents. Existing heading-based retrieval (Jeong et al., 2025) requires multiple LLM calls per document and returns sub-chunks within matched sections. We introduce ToC-guided page retrieval, which infers headings from visual formatting without LLM calls, embeds them as a parallel index, and loads full page sections. Across 1,280 conditions on 8 enterprise documents (5 to 195 pages), we find: (1) ToC is a significant main effect on answer quality (d = +0.41, p = 0.031), with the largest gains in completeness (+0.40) and usefulness (+0.40); (2) combined with answer-side verification, it outperforms query-side decomposition + verification (d = +0.32, p = 0.036); (3) ToC contributes 20% of citations despite adding only 2.9 pages per query; (4) gains are directionally larger on longer documents (up to +1.50 on 118 pages), though the trend does not reach significance with 8 documents; and (5) a 480-condition sensitivity analysis finds no significant parameter effects (all p > 0.38), confirming defaults are near-optimal. The contribution is both methodological (a new zero-LLM-cost retrieval algorithm) and empirical: factorial evidence that document-side, query-side, and answer-side enhancements are complementary, a three-way interaction not previously studied.
