LLM4Rec: A Comprehensive Survey on the Integration of Large Language Models in Recommender Systems—Approaches, Applications and Challenges
Sarama Shehmir, Rasha Kashef
Abstract
The synthesis of large language models (LLMs) and recommender systems has been a game-changer in tailored content onslaught with applications ranging from e-commerce, social media, and education to health care. This survey covers the usage of LLMs for content recommendations (LLM4Rec). LLM4Rec has opened up a whole set of challenges in terms of scale, real-time processing, and data privacy, all of which we touch upon along with potential future directions for research in areas such as multimodal recommendations and reinforcement learning for long-term engagement. This survey combines existing developments and outlines possible future developments, thus becoming a point of reference for other researchers and practitioners in developing the future of LLM-based recommendation systems.