RSPapers is a curated reading list of academic papers on recommender systems.
The repository addresses the challenge of navigating the vast literature on recommendation algorithms and techniques by organizing papers into thematic categories. Rather than providing implementations or tools, it serves as a structured guide to foundational and contemporary research, helping researchers and practitioners understand the landscape of recommender system approaches from multiple angles.
Developers and researchers building recommendation systems should use this list to ground their work in established theory and recent advances. The collection spans classical collaborative filtering methods through modern deep learning approaches, social recommendation techniques, and emerging areas like large language models for recommendation and agentic systems. It also covers specialized problem domains including cold start scenarios, point-of-interest recommendation, efficiency optimization, exploration-exploitation tradeoffs, explainability, privacy preservation, and click-through rate prediction. This breadth makes it suitable for anyone seeking to understand what has been tried in recommendation systems or looking to identify relevant prior work for a specific recommendation challenge.
The project maintains active curation with recent additions of sections on agentic recommender systems, large language models for recommendation, privacy and security in recommendation, and formal tutorials. The repository continues to expand its coverage of emerging topics while preserving its core organization around established problem areas and methodological approaches in the field.
