Tag: recommendation systems
Browse our exclusive articles!
Evaluating LLM-Initialized Bandits: Warm-Start vs Cold-Start
Explore how LLM-initialized bandits perform under noise and misalignment, revealing when warm-starting improves or harms algorithm efficiency.
Mitigating Popularity Bias in Personalized Recommendations
Discover how SPREE and Popularity Quantile Calibration improve user alignment and reduce popularity bias in recommender systems for better personalization.
UniMixer: Unified Scaling Architecture for Recommendation Systems
Discover UniMixer, a unified architecture boosting scaling efficiency and performance in recommendation systems with optimized token mixing.
MemRerank: Enhanced Personalized Product Reranking
MemRerank boosts e-commerce personalization by distilling user purchase history into concise signals for accurate product reranking and better recommendati...
GISTBench: Benchmarking LLMs for User Interest Verification
GISTBench evaluates LLMs' ability to verify user interests using novel metrics, enhancing personalization in recommendation systems with reliable datasets.