Researchers have developed RECAP, a framework designed to optimize streaming semantic user profiles for short-video recommendation systems. This closed-loop system uses LLM-based semantic updates and feedback from implicit user behavior to incrementally update user profiles within limited capacity. Experiments on Kuaishou data demonstrated improvements in recommendation metrics, and an online A/B test showed a statistically significant increase in average application usage time per user. AI
IMPACT Enhances recommendation system performance by improving user profile generation and feedback loops.
RANK_REASON Academic paper detailing a new framework for optimizing recommendation systems. [lever_c_demoted from research: ic=1 ai=1.0]
Read on arXiv cs.IR (Information Retrieval) →
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
