Researchers have developed a novel method called DEGR (Dual Exploration-Driven Generative Re-Ranking) to improve recommendation systems, particularly in scenarios with lower-quality upstream supply. DEGR employs a hybrid supervised-reinforcement learning approach, guided by an exploratory reward model that dynamically balances immediate user value with the potential for future discovery. This adaptive strategy aims to enhance user engagement and conversion rates by acting as a contextual bridge across different user requests. Experiments on the JD E-commerce recommendation system demonstrated DEGR's superiority over existing methods, yielding significant improvements in UCTR and PV metrics. AI
IMPACT Enhances recommendation system effectiveness by balancing immediate value with exploratory potential, potentially improving user engagement and conversion rates.
RANK_REASON Academic paper detailing a new method for recommendation systems. [lever_c_demoted from research: ic=1 ai=1.0]
Read on arXiv cs.IR (Information Retrieval) →
- alphaXiv
- arXiv
- CatalyzeX Code Finder for Papers
- CORE Recommender
- DagsHub
- Gotit.pub
- Hugging Face
- Influence Flower
- JD E-commerce recommendation system
- ScienceCast
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
