AIBullisharXiv – CS AI · 7h ago6/10
🧠
Breaking Information Cocoons: A Hyperbolic Framework for Balancing Exploration and Exploitation in Recommender Systems
Researchers propose HERec, a hyperbolic-geometry-based recommender system framework that balances content exploration and exploitation while mitigating information cocoons. The system combines semantic-enhanced hierarchical mechanisms with automatic clustering to improve diversity by 11.39% and utility by 5.49% over existing approaches.