Research

Paper

AI LLM February 27, 2026

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

Authors

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

Abstract

Existing scientific document retrieval (SDR) methods primarily rely on document-centric representations learned from inter-document relationships for document-document (doc-doc) retrieval. However, the rise of LLMs and RAG has shifted SDR toward question-driven retrieval, where documents are retrieved in response to natural-language questions (q-doc). This change has led to systematic mismatches between document-centric models and question-driven retrieval, including (1) input granularity (long documents vs. short questions), (2) semantic focus (scientific discourse structure vs. specific question intent), and (3) training signals (citation-based similarity vs. question-oriented relevance). To this end, we propose UniFAR, a Unified Facet-Aware Retrieval framework to jointly support doc-doc and q-doc SDR within a single architecture. UniFAR reconciles granularity differences through adaptive multi-granularity aggregation, aligns document structure with question intent via learnable facet anchors, and unifies doc-doc and q-doc supervision through joint training. Experimental results show that UniFAR consistently outperforms prior methods across multiple retrieval tasks and base models, confirming its effectiveness and generality.

Metadata

arXiv ID: 2602.23766
Provider: ARXIV
Primary Category: cs.IR
Published: 2026-02-27
Fetched: 2026-03-02 06:04

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2602.23766v1</id>\n    <title>UniFAR: A Unified Facet-Aware Retrieval Framework for Scientific Documents</title>\n    <updated>2026-02-27T07:44:02Z</updated>\n    <link href='https://arxiv.org/abs/2602.23766v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2602.23766v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>Existing scientific document retrieval (SDR) methods primarily rely on document-centric representations learned from inter-document relationships for document-document (doc-doc) retrieval. However, the rise of LLMs and RAG has shifted SDR toward question-driven retrieval, where documents are retrieved in response to natural-language questions (q-doc). This change has led to systematic mismatches between document-centric models and question-driven retrieval, including (1) input granularity (long documents vs. short questions), (2) semantic focus (scientific discourse structure vs. specific question intent), and (3) training signals (citation-based similarity vs. question-oriented relevance). To this end, we propose UniFAR, a Unified Facet-Aware Retrieval framework to jointly support doc-doc and q-doc SDR within a single architecture. UniFAR reconciles granularity differences through adaptive multi-granularity aggregation, aligns document structure with question intent via learnable facet anchors, and unifies doc-doc and q-doc supervision through joint training. Experimental results show that UniFAR consistently outperforms prior methods across multiple retrieval tasks and base models, confirming its effectiveness and generality.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.IR'/>\n    <published>2026-02-27T07:44:02Z</published>\n    <arxiv:primary_category term='cs.IR'/>\n    <author>\n      <name>Zheng Dou</name>\n    </author>\n    <author>\n      <name>Zhao Zhang</name>\n    </author>\n    <author>\n      <name>Deqing Wang</name>\n    </author>\n    <author>\n      <name>Yikun Ban</name>\n    </author>\n    <author>\n      <name>Fuzhen Zhuang</name>\n    </author>\n  </entry>"
}