Top-One Recommendation with Anonymous User Behaviors (Student Abstract)
Abstract
Top-one recommendation with anonymous user behaviors, also known as session-based recommendation (SBR), faces challenges of top-one ranking and short anonymous sequences. To this end, we propose a novel objective that combines (1) a reciprocal rank loss to directly optimize the benchmark metric of top-one recommendation, with (2) a listwise contrastive loss to handle short sequences through listwise augmented consistency regularization. Empirical studies demonstrate that optimizing the proposed objective significantly improves the performance of existing SBR baselines.
Cite
Text
Lu and Wu. "Top-One Recommendation with Anonymous User Behaviors (Student Abstract)." AAAI Conference on Artificial Intelligence, 2025. doi:10.1609/AAAI.V39I28.35274Markdown
[Lu and Wu. "Top-One Recommendation with Anonymous User Behaviors (Student Abstract)." AAAI Conference on Artificial Intelligence, 2025.](https://mlanthology.org/aaai/2025/lu2025aaai-top/) doi:10.1609/AAAI.V39I28.35274BibTeX
@inproceedings{lu2025aaai-top,
title = {{Top-One Recommendation with Anonymous User Behaviors (Student Abstract)}},
author = {Lu, Xiangkui and Wu, Jun},
booktitle = {AAAI Conference on Artificial Intelligence},
year = {2025},
pages = {29423-29425},
doi = {10.1609/AAAI.V39I28.35274},
url = {https://mlanthology.org/aaai/2025/lu2025aaai-top/}
}