TY - GEN
T1 - Private Set Operations from Circuit-Based PSI
AU - Kim, Jiseung
AU - Lee, Hyung Tae
AU - Son, Yongha
N1 - Publisher Copyright:
© 2026 Copyright held by the owner/author(s).
PY - 2026/6/9
Y1 - 2026/6/9
N2 - Private set operations (PSOs) have been extensively studied due to their wide range of applications. In this work, we present a novel framework for constructing advanced PSO protocols by treating circuit-based private set intersection (circuit-PSI) as a block-box building block, complemented by a minimal OT-based postprocessing layer customized for each target functionality. Our framework supports a variety of extensions, including PSI-Sum, which computes the sum of associated values; PSI-Card, which reveals the intersection cardinality; PSI-InnerProd, which computes the inner product over intersecting elements; and a private set union (PSU) protocol-all without leaking intermediate information, a vulnerability known as during-execution leakage. We implement and benchmark our protocols against closely related PSO protocols. The results show that our constructions outperform state-of-the-art protocols in total runtime. An important additional advantage of our approach is the ability to offload expensive OT operations to the setup phase, significantly improving online efficiency. For instance, in terms of online execution time under LAN settings, PSI-Card and PSI-Sum achieve 9.92 - 13× speedups, PSU achieves 10.8 - 11.7×, and PSI-InnerProd demonstrates up to a remarkable 47 - 56× improvement over prior work with comparable communication overhead.
AB - Private set operations (PSOs) have been extensively studied due to their wide range of applications. In this work, we present a novel framework for constructing advanced PSO protocols by treating circuit-based private set intersection (circuit-PSI) as a block-box building block, complemented by a minimal OT-based postprocessing layer customized for each target functionality. Our framework supports a variety of extensions, including PSI-Sum, which computes the sum of associated values; PSI-Card, which reveals the intersection cardinality; PSI-InnerProd, which computes the inner product over intersecting elements; and a private set union (PSU) protocol-all without leaking intermediate information, a vulnerability known as during-execution leakage. We implement and benchmark our protocols against closely related PSO protocols. The results show that our constructions outperform state-of-the-art protocols in total runtime. An important additional advantage of our approach is the ability to offload expensive OT operations to the setup phase, significantly improving online efficiency. For instance, in terms of online execution time under LAN settings, PSI-Card and PSI-Sum achieve 9.92 - 13× speedups, PSU achieves 10.8 - 11.7×, and PSI-InnerProd demonstrates up to a remarkable 47 - 56× improvement over prior work with comparable communication overhead.
KW - circuit-PSI
KW - private set operations
KW - private set union
UR - https://www.scopus.com/pages/publications/105042691929
U2 - 10.1145/3748522.3779856
DO - 10.1145/3748522.3779856
M3 - Conference paper
AN - SCOPUS:105042691929
T3 - Proceedings of the ACM Symposium on Applied Computing
SP - 1857
EP - 1866
BT - SAC 2026 - 41st Annual ACM Symposium on Applied Computing
PB - Association for Computing Machinery
T2 - 41st Annual ACM Symposium on Applied Computing, SAC 2026
Y2 - 23 March 2026 through 27 March 2026
ER -