TME-PSR: Time-aware, Multi-interest, and Explanation Personalization for Sequential Recommendation
Published in arXiv 2026, 2026
TME-PSR integrates time-aware, multi-interest, and explanation personalization for sequential recommendation. A dual-view gated time encoder captures individual temporal rhythms for recommendation and explanation. A lightweight multihead Linear Recurrent Unit (LRU) architecture models fine-grained sub-interests efficiently.
A dynamic dual-branch mutual information weighting mechanism adapts the semantic alignment between recommendations and explanations to each user. Experiments on three real-world datasets show improved recommendation accuracy and explanation quality with lower computational cost than the evaluated baselines.
Recommended citation: Qingzhuo Wang, Leilei Wen, Juntao Chen, Kunyu Peng, Ruiyang Qin, Zhihua Wei, Wen Shen. (2026). "TME-PSR: Time-aware, Multi-interest, and Explanation Personalization for Sequential Recommendation." arXiv 2026.
Download Paper