ברכה שפירא

User preferences, intents, behaviors, and contexts evolve over time, making temporal reasoning a fundamental challenge in recommender systems. Accurately capturing both short-term and long-term dynamics is essential for improving personalization and recommendation quality across domains such as e-commerce, media consumption, mobility, travel, and finance. In parallel, recent advances in large language models (LLMs) and foundation models create new opportunities for incorporating sequential interaction histories, evolving preferences, and time-dependent features into next-generation recommendation pipelines. RecTemp 2026 provides a dedicated forum for researchers and practitioners interested in modeling, analyzing, and leveraging temporal information in recommender systems. The workshop addresses topics including sequential and session-based recommendation, time-aware learning, evolving user profiles, temporal context integration, cross-domain temporal patterns, and temporal reasoning in LLM-based recommender systems.