Q2EI: Query-to-Entity Inference for Semantic Condensation in Domain-Specific Retrieval

Yixuan Sun, Zhenqin Xu, HanFeng Zhai, Zishu Yu, Xiaohui Peng


Abstract
Retrieval-Augmented Generation (RAG) remains unreliable in specialized domains due to semantic and lexical mismatch between lay queries and professional terminology, and existing generative expansion often introduces redundancy or hallucinations that cause semantic drift. We propose Generative Query Condensation (GQC), a query rewriting strategy that reframes rewriting as semantic condensation rather than expansion. To operationalize GQC, we introduce Query-to-Entity Inference (Q2EI), an entity-centric rewriting method that realizes semantic condensation through explicit inference of the underlying target entity. By moving semantic alignment from retrieval-time vector matching to the rewriting stage, Q2EI produces information-dense query representations. Experimental results on medical and legal benchmarks show that Q2EI consistently outperforms strong baselines across retrievers, improving retrieval effectiveness while substantially reducing rewriting token consumption compared to generative expansion methods. Further analysis confirms that these gains primarily arise from accurate entity inference, and that Q2EI’s semantic condensation design limits error amplification when inference is imperfect, leading to more stable and interpretable retrieval behavior.
Anthology ID:
2026.findings-acl.1791
Volume:
Findings of the Association for Computational Linguistics: ACL 2026
Month:
July
Year:
2026
Address:
San Diego, California, United States
Editors:
Maria Liakata, Viviane P. Moreira, Jiajun Zhang, David Jurgens
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
35943–35958
Language:
URL:
https://aclanthology.org/2026.findings-acl.1791/
DOI:
10.18653/v1/2026.findings-acl.1791
Bibkey:
Cite (ACL):
Yixuan Sun, Zhenqin Xu, HanFeng Zhai, Zishu Yu, and Xiaohui Peng. 2026. Q2EI: Query-to-Entity Inference for Semantic Condensation in Domain-Specific Retrieval. In Findings of the Association for Computational Linguistics: ACL 2026, pages 35943–35958, San Diego, California, United States. Association for Computational Linguistics.
Cite (Informal):
Q2EI: Query-to-Entity Inference for Semantic Condensation in Domain-Specific Retrieval (Sun et al., Findings 2026)
Copy Citation:
PDF:
https://aclanthology.org/2026.findings-acl.1791.pdf
Checklist:
 2026.findings-acl.1791.checklist.pdf