Researchers have developed HGenPush, a novel generative recommendation architecture designed for industrial push notification systems. This system addresses the limitations of existing methods by integrating both content and author recommendations within a unified framework. HGenPush utilizes a hybrid user behavior understanding module and a lightweight multi-token prediction method to improve efficiency and user experience. When deployed on Kuaishou, a large-scale short-video platform, HGenPush demonstrated a 0.181% increase in daily active users. AI
IMPACT This architecture could improve user engagement and retention on content platforms by providing more tailored recommendations.
RANK_REASON The item is an academic paper describing a new architecture and its performance on a specific platform. [lever_c_demoted from research: ic=1 ai=1.0]
Read on arXiv cs.IR (Information Retrieval) →
- alphaXiv
- arXiv
- CatalyzeX Code Finder for Papers
- Connected Papers
- CORE Recommender
- DagsHub
- Gotit.pub
- HGenPush
- Hugging Face
- Influence Flower
- Kuaishou
- Litmaps
- ScienceCast
- scite Smart Citations
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
