Dynamic Injection of Entity Knowledge into Dense Retrievers
Abstract
Dense retrievers often struggle with queries involving less-frequent entities due to their limited entity knowledge. We propose the Knowledgeable Passage Retriever (KPR), a BERT-based retriever enhanced with a context-entity attention layer and dynamically updatable entity embeddings. This design enables KPR to incorporate external entity knowledge without retraining. Experiments on three datasets demonstrate that KPR consistently improves retrieval accuracy, with particularly large gains on the EntityQuestions dataset. When built on the off-the-shelf bge-base retriever, KPR achieves state-of-the-art performance among similarly sized models on two datasets. Models and code are released at https://github.com/knowledgeable-embedding/knowledgeable-embedding.
Cite
@article{arxiv.2507.03922,
title = {Dynamic Injection of Entity Knowledge into Dense Retrievers},
author = {Ikuya Yamada and Ryokan Ri and Takeshi Kojima and Yusuke Iwasawa and Yutaka Matsuo},
journal= {arXiv preprint arXiv:2507.03922},
year = {2025}
}
Comments
EMNLP Findings