November 14, 2026 (see also the program)
The workshop on Trustworthy Recommender Systems across Learning, Relearning, and Unlearning (TrustRS) is a half-day workshop held in conjunction with IEEE ICDM 2026.
Recommender systems have become critical infrastructure shaping decisions across e-commerce, social media,healthcare, and education. While conventional research prioritizes accuracy, the growing societal impact of recommender systems demands greater attention to trust, which encompasses transparency, fairness, and privacy.
TrustRS examines trustworthiness through the lens of a recommender system's full operational lifecycle: Learning(models discover user preferences from training data), Relearning (models adapt recommendations to new user behaviors or shifting environments), and Unlearning (models selectively remove sensitive or outdated user data to comply with privacy regulations or correct undesirable recommendation behaviors). Each stage introduces distinct trust challenges that warrant dedicated investigation.
As data privacy regulations such as GDPR and CCPA evolve, and AI accountability becomes a societal imperative, TrustRS closely aligns with ICDM's commitment to responsible AI by advancing research at the intersection of recommender systems, data privacy, and algorithmic fairness.
The workshop will be held in conjunction with the IEEE International Conference on Data Mining (ICDM) 2026 on Saturday, November 14, 2026, as a half-day workshop.