This paper surveys the evolving field of autonomous information-seeking within recommender systems, driven by the integration of large language model-based agents. It introduces a taxonomy based on autonomy levels and three core paradigms: agent-assisted recommendation, agent-as-recommender, and agent-as-user-simulator. The survey also examines current evaluation methodologies, discusses their limitations, and outlines open challenges in areas such as lifelong user modeling, trustworthiness, and efficiency for developing more human-aligned recommendation agents. AI
IMPACT This survey provides a framework for understanding and developing more autonomous and human-aligned AI agents within recommender systems.
RANK_REASON The item is a survey paper published on arXiv, detailing research on AI agents in recommender systems. [lever_c_demoted from research: ic=1 ai=1.0]
Read on Hugging Face Daily Papers →
- agentic recommender system
- Autonomous Information Seeking: A Roadmap for Agentic Recommender Systems
- Hugging Face
- AI agent
- large language model
- recommender system
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
