Nov 2026· Proceedings of the 35th ACM International Conference on Information and Knowledge Management· pp. 8818-8826· 0 citations
Abstract
Temporal distribution shift (TDS) degrades the generalization of recommender systems: models trained on past interactions face systematically different user behavior in future time windows. While incremental training on recent data is widely used in industry production, it can over-adapt to transient patterns and exacerbate instability. Existing approaches such as invariant learning and self-supervised learning offer partial solutions but often suffer from unstable temporal generalization, representation collapse, or inefficient data utilization. To address these limitations, we propose ELBOTDS, a probabilistic framework for temporally robust recommendation. First, we identify key shifting factors on real-world production data and design a simple yet effective data augmentation strategy that simulate realistic time-varying factors to extend the training support. Second, we derive a self-supervised variational objective ELBOTDS to harness the benefits of this extended distribution while preventing representation collapse. Extensive experiments on both a public benchmark and a large-scale industrial dataset demonstrate ELBOTDS consistently improves next-period ranking performance over strong baselines, yielding a 2.33% uplift in GMV per user online and has been successfully deployed in production. Code is available at https://github.com/FuCongResearchSquad/ELBO4TDS.
DynFreqRec, a Frequency-Aware Training–Inference Synergistic Enhancement Framework for sequential recommendation, builds on an existing Base Time-Frequency Encoder that uses adaptive frequency-domain filtering and the Haar Wavelet Transform to characterize global preferences, periodic interests, and local non-stationar...
Training large-scale recommender systems on dynamic industrial platforms, such as Short-Form Video (SFV) services with billions of daily interactions, faces critical stability challenges. Rapid distribution shifts in user data and increasing model complexity frequently trigger a divergence cascade, typically initiating...
Yue-Ning Li, Ze-Yu Wang, Zhen Zhang et al.· 0 citations
This work proposes an innovative meta-learning-driven conditional diffusion model that uses a generative model to directly model the data distribution and learn personalized user patterns, enhancing recommendation stability and the meta-learning framework implicitly constrains the optimization process of the diffusion...
Di Jin, Run-Ze Li, Jia-Qi Cui et al.· ACM Transactions on Informat...· 0 citations
Sequential recommender systems struggle to simultaneously capture persistent long-term preferences and volatile short-term intentions, while existing dual-stream architectures are computationally inefficient when handling long sequences and suffer from feature entanglement between the two interest streams. To address t...
Yue Xie, Yan Zhang, Guo Chen et al.· Proceedings of the 35th ACM...· 0 citations
Long-sequence modeling is increasingly important in recommender systems for capturing users’ evolving and long-term interests. In advertising, however, user interaction histories are often highly sparse due to limited exposure opportunities, making ad-only behavior sequences insufficient for effective long-sequence rec...
Xian Hu, Ming Yue, Zhi-Xiang Feng et al.· Proceedings of the 20th ACM...· 0 citations
Personalized product recommendation is a central challenge in online platforms, where accurately capturing user preferences and adapting to dynamic behavior patterns directly impacts conversion rates and user satisfaction. Traditional collaborative filtering and content-based methods struggle with data sparsity, cold-s...
Yu-Lu Zhou· International Conference on...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.