CARS - Workshop on Context-Aware Recommender Systems
Contextual information has long been recognized as a fundamental modeling dimension in both social science and technological disciplines, and its importance continues to grow as recommender systems strive to deliver more relevant and personalized results. Over the years, several workshops on Context-Aware Recommender Systems (CARS) have explored the integration of contextual information into traditional recommendation frameworks. Much of this work has followed the so-called representational view, where context is modeled using predefined and static factors such as time, location, or device. In recent years, however, the field has undergone a significant transformation. Advances in generative AI and foundation models, particularly Large Language Models (LLMs), have enabled new ways to infer context directly from unstructured and multimodal data sources. Modern recommender systems increasingly move beyond explicitly defined context toward latent, inferred, and dynamic contextual representations, extracted from signals such as text, images, interaction sequences, and environmental data . This shift opens new opportunities for understanding user intent, modeling evolving preferences, and improving recommendation quality in complex, real-world environments.
Alongside these developments, novel approaches such as sequence-aware recommendation, intent-aware systems, and latent context modeling have emerged, allowing systems to capture both short- and long-term user behavior. At the same time, the growing deployment of recommender systems in mobile, IoT, conversational, and multi-stakeholder environments introduces new challenges. These include the need to model partially observable and rapidly changing contexts, while also addressing critical concerns such as privacy, security, fairness, and transparency.
The CARS 2026 workshop aims to rethink the role of contextual information in recommender systems in this new era of AI-driven personalization. In particular, the workshop will focus on how contextual signals and user intent can be inferred from multimodal data, how these signals can be integrated into next-generation recommendation architectures, and how systems can remain explainable, adaptive, and privacy-preserving. The workshop seeks to foster cross-disciplinary discussion, identify key research challenges, and advance innovative approaches for building trustworthy and effective context-aware recommender systems.
The CARS 2026 workshop will be held in person on September 28th, 2026, in the morning (8:30 am - 12:30 pm), and will be co-located with RecSys’26 in Minneapolis, Minnesota, USA.
