arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2606.09459cs.CL
收藏

AbstRAG:面向检索问题的抽象学习

AbstRAG: Learning to Abstract for Retrieval Problems

Idiap Research Institute(Idiap 研究所);École Polytechnique Fédérale de Lausanne (EPFL)(洛桑联邦理工学院 (EPFL));São Paulo State University(圣保罗州立大学);University of Manchester(曼彻斯特大学);CRUK National Biomarker Centre(英国癌症研究中心国家生物标志物中心)

Lei Xu, Xin Quan, Daniel Pedronette, André Freitas

AI总结:

针对查询与文档证据间的抽象鸿沟问题,提出AbstRAG方法,通过将抽象作为显式检索对象,并采用反思性精炼机制,在三个基准上提升了检索和生成性能。

摘要:

当查询、文档证据和用户意图以不同抽象级别表达时,检索增强生成常常失败。查询可能询问一个类别、关系或事件,而文档仅陈述具体实例、间接框架或限定表述。我们将这种不匹配定义为抽象鸿沟:将查询意图与可用证据对齐所需的最小类型假设集合。为弥合这一鸿沟,我们引入AbstRAG,将抽象视为显式检索对象。AbstRAG将查询-证据鸿沟分解为表达、概念、意图-证据和事件类型组件,并通过结合匹配质量、查询无关的效用先验以及所需桥梁的成本来评分相关性。其核心机制是反思性精炼:批评者诊断检索失败,定位失败的抽象操作符,提出最小的阶段特定补丁,并仅在充分性和压缩控制下接受补丁。在三个文档内检索基准上与七个基线对比,AbstRAG在21个配对自助法对比中的18个上以nDCG@10胜出,并在三个基准上分别将生成准确率提升1.9%、5.2%和4.0%;消融实验证实,反思性精炼驱动了大部分检索增益,而仅压缩控制就在压力切片上将过度扩展假阳性从73.7%降至0%。

英文摘要:

Retrieval-augmented generation often fails when the query, the document evidence, and the user's intent are expressed at different levels of abstraction. A query may ask about a class, a relation, or an event, while the document only states specific instances, indirect framings, or scoped formulations. We define this mismatch as an abstraction gap: the minimal set of typed assumptions required to align query intent with the available evidence. To close this gap, we introduce AbstRAG, which treats abstraction as an explicit retrieval object. AbstRAG decomposes the query--evidence gap into expression, conceptual, intent--evidence, and event-type components, and scores relevance by combining match quality, a query-independent utility prior, and the cost of the required bridges. Its central mechanism is reflective refinement: a critic diagnoses retrieval failures, localizes the failed abstraction operator, proposes a minimal stage-specific patch, and accepts the patch only under sufficiency and compression controls. Across three within-document retrieval benchmarks against seven baselines, AbstRAG outperforms on nDCG@10 in 18 of 21 paired-bootstrap contrasts and improves generation accuracy by 1.9%, 5.2%, and 4.0% across the three benchmarks; ablations confirm that reflective refinement drives most of the retrieval gain and the compression control alone reduces over-expansion false positives from 73.7% to 0% on a stress slice.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑