解决文本中聚合查询的完整性难题，提出新基准与可解释方法
Aggregation Queries over Unstructured Text: Benchmark and Agentic Method
- 将聚合查询分解为消歧-过滤-聚合三阶段，提升可解释性
- 在大规模语料上验证，相比现有方法证据覆盖率达91.3%
- 适合需要高完整性的信息检索与智能问答系统研究者
自由文本上的聚合查询是一个长期存在但研究不足的问题。与普通问答不同，聚合查询需收集全部证据，系统必须‘找全’而非‘找一个’。现有范式如Text-to-SQL和检索增强生成难以保证完整性。本文在语料限定条件下形式化实体级聚合查询，并提出AGGBench基准，用于评估真实大规模语料下的完整性导向聚合能力。为此，我们设计DFA（消歧-过滤-聚合）模块化代理基线，将聚合查询分解为可解释阶段，暴露模糊性、过滤和聚合中的关键失败模式。实证结果表明，DFA在证据覆盖率上持续优于强RAG和代理基线，在多个数据集上平均提升8.7个百分点。
原文摘要 · Abstract (English)
Aggregation query over free text is a long-standing yet underexplored problem. Unlike ordinary question answering, aggregate queries require exhaustive evidence collection and systems are required to "find all," not merely "find one." Existing paradigms such as Text-to-SQL and Retrieval-Augmented Generation fail to achieve this completeness. In this work, we formalize entity-level aggregation querying over text in a corpus-bounded setting with strict completeness requirement. To enable principled evaluation, we introduce AGGBench, a benchmark designed to evaluate completeness-oriented aggregation under realistic large-scale corpus. To accompany the benchmark, we propose DFA (Disambiguation--Filtering--Aggregation), a modular agentic baseline that decomposes aggregation querying into interpretable stages and exposes key failure modes related to ambiguity, filtering, and aggregation. Empirical results show that DFA consistently improves aggregation evidence coverage over strong RAG and agentic baselines. The data and code are available in \href{https://anonymous.4open.science/r/DFA-A4C1}.
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。