LLM-based-Agent-for-RS

This repository collects papers related to LLM-based Agent for Recommender Systems.

Existing work can be roughly divided into four categories: User Agent, Item Agent, User-Item Collabration, and Recommender Agent.

• Background

• Category

  • User Agent
  • Item Agent
  • User-Item Collabration
  • Recommender Agent

Background

LLM-based Agent, or called LLM-empowered generative agents, refers to autonomous agents which employs LLMs as brains or central controllers, aiming at leveraging the human-like capabilities of LLMs.

In recommender systems, LLM-based Agent can be applied for user simulation, better recommendation, etc.

Category

User Agent

• RecAgent - When Large Language Model based Agent Meets User Behavior Analysis: A Novel User Simulation Paradigms [Paper]
• Agent4Rec - On Generative Agents in Recommendation [Paper]

Item Agent

As far as we know, there is currently no related work that only builds item agents.

User-Item Collabration

• AgentCF - AgentCF: Collaborative Learning with Autonomous Language Agents for Recommender Systems [Paper]

Recommender Agent

• RecMind - RecMind: Large Language Model Powered Agent For Recommendation [Paper]
• InteRecAgent - Recommender AI Agent: Integrating Large Language Models for Interactive Recommendations [Paper]
• RAH - RAH! RecSys-Assistant-Human: A Human-Centered Recommendation Framework with LLM Agents [Paper]