Performance analysis of continuous-variable quantum key distribution (CVQKD) has been one of the focuses of quantum communications. In this paper, we propose an approach to enhancing the secret rate of CVQKD with the multi-core fiber (MCF) system that transmits multiple spatial modes simultaneously. The excess noise contributed by the inter-core crosstalk between cores can be effectively suppressed by quantum channel wavelength management, leading to the performance improvement of the MCF-based CVQKD system. In the security analysis, we perform numerical simulations for the Gaussian-modulated coherent state CVQKD protocol, considering simultaneously the extra insert loss of fan-in/fan-out (FIFO), which is the extra optical device that should be used at the input and the output of the fiber. Simulation results show that the performance of the one-way and two-way protocols for each core are slightly degraded because of the insert loss of the FIFO, but the total secret key rate can be increased, whereas the performance of the measurement-device-independent CVQKD protocol will be degraded due to the effect of the insert loss of the FIFO. These results may provide theoretical foundation for the space-division multiplexing CVQKD system.

Traditional book recommendation methods predominantly rely on collaborative filtering and context-based approaches. However, existing methods fail to account for the order of users’ book borrowings and the duration they hold them, both of which are crucial indicators reflecting users’ book preferences. To address this challenge, we propose a book recommendation framework called DPBD, which disentangles preferences based on borrowing duration, thereby explicitly modeling temporal patterns in library borrowing behaviors. The DPBD model adopts a dual-path neural architecture comprising the following: (1) The item-level path utilizes self-attention networks to encode historical borrowing sequences while incorporating borrowing duration as an adaptive weighting mechanism for attention score refinement. (2) The feature-level path employs gated fusion modules to effectively aggregate multi-source item attributes (e.g., category and title), followed by self-attention networks to model feature transition patterns. The framework subsequently combines both path representations through fully connected layers to generate user preference embeddings for next-book recommendation. Extensive experiments conducted on two real-world university library datasets demonstrate the superior performance of the proposed DPBD model compared with baseline methods. Specifically, the model achieved 13.67% and 15.75% on HR@1 and 15.75% and 12.90% on NDCG@1 across the two datasets.