Tutorial: Continual Recommender Systems

This tutorial has been accepted and will be presented as part of CIKM 2025 in November 2025.

Abstract

Modern recommender systems operate in uniquely dynamic settings: user interests, item pools, and popularity trends shift continuously, and models must adapt in real time without forgetting past preferences. While existing tutorials on continual or lifelong learning cover broad machine learning domains (e.g., vision and graphs), they do not address recommendation-specific demands—such as balancing stability and plasticity per user, handling cold-start items, and optimizing recommendation metrics under streaming feedback. This tutorial aims to make a timely contribution by filling that gap.

We begin by reviewing the background and problem settings, followed by a comprehensive overview of existing approaches, including replay-based and regularization-based methods. We then highlight recent efforts to apply continual learning to practical deployment environments, such as resource-constrained systems and sequential interaction settings. Finally, we discuss open challenges and future research directions. We believe this tutorial will be valuable to researchers and practitioners in recommender systems, data mining, and artificial intelligence, and will benefit a wide range of real-world application domains related to information retrieval.

The audience is expected to have basic knowledge of probability, linear algebra, and machine learning, but no prior familiarity with specific continual learning algorithms is required. 

Outline and Timeline

This tutorial is designed to run for a half day, totaling approximately 3 hours including short breaks. It will be organized into the following parts:

• Part I: Introduction and Background (30 minutes)

  • Problem definitions and settings

  • Key challenges

  • Applications and use cases

• Part II: Experience-Replay-based Methods (35 min)

  • Sample selection for experience replay

  • Replay-based model enhancement

• Part III: Regularization-based Methods (35 min)

  • What knowledge to regularize

  • Which temporal knowledge to regularize

  • Personalization of regularization

• Part IV: Beyond Traditional Settings (35 min)

  • Resource-constrained environments

  • Sequential interaction environments

• Part V: Open Challenges and Future Directions (35 min)

  • Trustworthiness

  • Adaptation to Foundational Models

  • Unified Models for Recommendation and Search

Slides

Incoming Soon!

Speakers

The presenters and contributors of this tutorial are Hyunsik Yoo, SeongKu Kang, and Hanghang Tong. Their biographies and areas of expertise are provided below.

References

This list includes the prior works covered in this tutorial:

• Ahrabian, K., Xu, Y., Zhang, Y., Wu, J., Wang, Y., & Coates, M. (2021, October). Structure aware experience replay for incremental learning in graph-based recommender systems. In Proceedings of the 30th ACM International Conference on Information & Knowledge Management (pp. 2832-2836).

• Mi, F., Lin, X., & Faltings, B. (2020, September). Ader: Adaptively distilled exemplar replay towards continual learning for session-based recommendation. In Proceedings of the 14th ACM Conference on Recommender Systems (pp. 408-413).

• Cai, G., Zhu, J., Dai, Q., Dong, Z., He, X., Tang, R., & Zhang, R. (2022, July). Reloop: A self-correction continual learning loop for recommender systems. In Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval (pp. 2692-2697).

• Zhu, J., Cai, G., Huang, J., Dong, Z., Tang, R., & Zhang, W. (2023, August). ReLoop2: Building Self-Adaptive Recommendation Models via Responsive Error Compensation Loop. In Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (pp. 5728-5738).

• Zhang, X., Chen, Y., Ma, C., Fang, Y., & King, I. (2024, March). Influential exemplar replay for incremental learning in recommender systems. In Proceedings of the AAAI Conference on Artificial Intelligence (Vol. 38, No. 8, pp. 9368-9376).

• Qin, J., Liu, W., Zhang, W., & Yu, Y. (2025, April). D2K: Turning historical data into retrievable knowledge for recommender systems. In Proceedings of the ACM on Web Conference 2025 (pp. 472-482).

• Wang, Y., Zhang, Y., Valkanas, A., Tang, R., Ma, C., Hao, J., & Coates, M. (2023, June). Structure aware incremental learning with personalized imitation weights for recommender systems. In Proceedings of the AAAI Conference on Artificial Intelligence (Vol. 37, No. 4, pp. 4711-4719).

• Xu, Yishi, et al. "Graphsail: Graph structure aware incremental learning for recommender systems." Proceedings of the 29th ACM International Conference on Information & Knowledge Management. 2020.

• Wang, Y., Zhang, Y., & Coates, M. (2021, October). Graph structure aware contrastive knowledge distillation for incremental learning in recommender systems. In Proceedings of the 30th ACM International Conference on Information & Knowledge Management (pp. 3518-3522).

• Yoo, H., Kang, S., Qiu, R., Xu, C., Wang, F., & Tong, H. (2025). Embracing plasticity: Balancing stability and plasticity in continual recommender systems. In Proceedings of the 48th International ACM SIGIR Conference on Research and Development in Information Retrieval.

• Lee, G., Kang, S., Kweon, W., & Yu, H. (2024, August). Continual Collaborative Distillation for Recommender System. In Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (pp. 1495-1505).

• Lee, G., Yoo, H., Hwang, J., Kang, S., & Yu, H. (2025). Leveraging Historical and Current Interests for Continual Sequential Recommendation. arXiv preprint arXiv:2506.07466.

• Liu, L., Cai, L., Zhang, C., Zhao, X., Gao, J., Wang, W., ... & Li, Q. (2023, July). Linrec: Linear attention mechanism for long-term sequential recommender systems. In Proceedings of the 46th International ACM SIGIR Conference on Research and Development in Information Retrieval (pp. 289-299).