Topics

TrustRS invites researchers and practitioners to explore trust challenges spanning the full lifecycle of recommender systems. Topics of interest include, but are not limited to:

Learning

•  Fairness-aware and bias-mitigated recommendation

• Explainable and transparent recommendation models

• Privacy-preserving recommendation via federated learning or differential privacy

• Robustness of recommender systems against adversarial attacks and data poisoning

Relearning

• Continual and incremental learning for evolving user preferences

• Adaptive recommendation under distribution shift and concept drift

• Efficient model updating with fairness and privacy guarantees

Unlearning

• Machine unlearning for traditional recommendation and LLM-based recommendation

• Efficient approximate unlearning in graph-based and sequential recommender systems

• Unlearning verification and auditing in recommender systems

Instructions for Submissions 

Authors are invited to submit original papers related to the topics of TrustRS 2026. Each submission may contain a maximum of 8 pages plus 2 extra pages and will be peer-reviewed. 

All submissions should follow the IEEE 2-column format, consistent with the formatting requirement of the ICDM 2026 main conference. Please refer to the ICDM 2026 Main Conference Submission Guidelines for the formatting template. 

Accepted workshop papers will be published in the dedicated ICDMW proceedings by the IEEE Computer Society Press.  

Paper Submission Site 

The paper submission site will be announced soon.