An Enhanced Privacy-Preserving Recommender System (https://doi.org/10.1007/978-981-13-7561-3_18)
We modified the trusted server assumption from an honest server to semi-honest server, and improved the efficiency of an existing PPRS protocol. The empirical results show the proposed modification performs well and security is much better than original work.

Item-Based Privacy-Preserving Recommender System with Offline Users and Reduced Trust Requirements (https://doi.org/10.1007/978-3-030-36945-3_12)
The paper proposes PPRS scheme that does not require all the users to be available online while the server generates recommendations for them. This is significant as most of the existing PPRSs that use a partially homomorphic encryption approach require that all users remain online and contribute to the recommendation generation.

Faster Private Rating Update via Integer-Based Homomorphic Encryption (https://doi.org/10.1007/978-3-030-92571-0_15)
This work proposes a novel approach that allows users to update their ratings for an item on the remote server securely and efficiently. The protocol is proposed for PPRS as a use case but it has a wider applicability. Any system architecture, where users need to modify a memory block at the remote server, without letting the server know which memory block was modified, this protocol can be very helpful.

Efficient privacy preserving top-k recommendation using homomorphic sorting (https://eprint.iacr.org/2022/1564)
Leveraging the efficient Batcher's sort with secure bit decomposition protocol, we proposed an efficient way of sorting under encryption. The server does not learn anything about the input and the user gets a sorted vector. The architecture uses two servers namely, the keyserver and dataserver. None of the servers learn user data and any relationship among the data items in the input vector.