arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2602.23766cs.IR
收藏

UniFAR: 一种统一的面向面检索框架用于科学文档

UniFAR: A Unified Facet-Aware Retrieval Framework for Scientific Documents

School of Computer Science and Engineering, Beihang University(北京航空航天大学计算机学院);School of Artificial Intelligence, Beihang University(北京航空航天大学人工智能学院)

Zheng Dou, Zhao Zhang, Deqing Wang, Yikun Ban, Fuzhen Zhuang

AI总结:

UniFAR提出一种统一的面向面检索框架,通过自适应多粒度聚合、可学习的面向锚点和联合训练,有效解决文档与问题驱动检索之间的不匹配问题,提升科学文档检索性能。

摘要:

现有的科学文档检索(SDR)方法主要依赖于从文档间关系中学习的文档中心表示来进行文档-文档(doc-doc)检索。然而,随着LLM和RAG的兴起,SDR逐渐转向以问题为导向的检索,其中文档是针对自然语言问题(q-doc)进行检索的。这种转变导致了文档中心模型与问题驱动检索之间的系统性不匹配,包括(1)输入粒度(长文档 vs. 短问题),(2)语义重点(科学论述结构 vs. 具体问题意图),以及(3)训练信号(引用基于的相似性 vs. 问题导向的相关性)。为此,我们提出了UniFAR,一种统一的面向面检索框架,以在单一架构内联合支持doc-doc和q-doc SDR。UniFAR通过自适应多粒度聚合解决粒度差异,通过可学习的面向锚点对齐文档结构与问题意图,并通过联合训练统一doc-doc和q-doc监督。实验结果表明,UniFAR在多个检索任务和基础模型上均优于先前方法,证实了其有效性和通用性。

英文摘要:

Scientific document retrieval (SDR) plays a critical role in modern scientific research, supporting knowledge discovery and evidence-based reasoning. It has evolved along two paradigms: document--document (doc-doc) retrieval driven by inter-document contrastive learning, and question--document (q-doc) retrieval emerging from LLMs and RAG for natural-language interaction. In practice, scientific workflows rely on both paradigms, requiring retrieval of related papers given a seed document and identifying relevant documents given a user question. However, existing methods typically treat these paradigms separately, hindering their complementary strengths. To address this, we propose UniFAR, a unified facet-aware retrieval framework that jointly supports doc-doc and q-doc retrieval within a shared representation space. UniFAR introduces a multi-granularity representation and aggregation module to unify the encoding of short questions and long documents, and a facet-level modeling mechanism with learnable anchors to capture structured semantic roles and complex user intents. It further adopts a facet-aware joint training strategy that integrates doc-doc and q-doc contrastive objectives with facet-level alignment, enabling unified learning from both inter-document relations and question-oriented supervision. Experiments on three benchmark datasets under both doc-doc and q-doc settings across multiple backbone models show that UniFAR consistently outperforms strong baselines and generalizes effectively across different backbones.

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。

↑