Exploring the Potential of Dimension Reduction in Building Efficient Dense Retrieval Systems
Lists
Only you see this
Abstract
Dense retrievers utilize pretrained language models to encode queries and documents as high-dimensional embeddings for retrieval. Nevertheless, these high-dimensional embeddings usually result in more expensive index storage and higher retrieval latency. In this paper, we further explore the potential of building a lightweight dense retrieval system by combining the dimension reduction in the encoding-indexing pipeline. Our experiments demonstrate that the encoding-compression-indexing-retrieval method can conduct an efficient dense retrieval system, reducing the retrieval latency of 96%, while maintaining comparable retrieval effectiveness. Our further analyses illustrate that the dimensional reduction method can broaden its retrieval effectiveness in different domains and cooperate with different index building methods.
BibTeX
@inproceedings{xu2024exploring,
author = {Zhipeng Xu and Zhenghao Liu and Yu Gu and Ge Yu},
title = {{Exploring the Potential of Dimension Reduction in Building Efficient Dense Retrieval Systems}},
booktitle = {CCIR},
year = {2024},
doi = {10.1007/978-981-96-1710-4\_6},
}Loading related content...