Personalized Recommendation Tool Learning via Autonomous Language Agents

2026-07-22

AI Summary

This work addresses the performance bottlenecks of large language models (LLMs) in full-ranking recommendation tasks, which stem from hallucination and limited context length. To overcome these limitations, the authors propose a novel recommendation framework based on autonomous language agents. In this framework, an LLM serves as a central planner that dynamically invokes multiple traditional recommendation models for scoring, while leveraging user profiles to enable personalized tool selection. By incorporating memory-augmented and reflection mechanisms, the agent can perform user-aware dynamic evaluation and optimization of recommendation tools without modifying the underlying LLM. Experimental results on three public datasets demonstrate that the proposed approach significantly outperforms both conventional recommender systems and LLM-based baselines, achieving substantial improvements in full-ranking recommendation performance.

Abstract

Although large language models (LLMs) have recently gained traction in recommender systems due to their strong reasoning capabilities and extensive world knowledge, previous LLM-based agents suffer from hallucination and context-length limitations, and thus are not suitable for full-ranking recommendation tasks. To circumvent these limitations through architectural design rather than modifying the LLM itself, we propose an agent-based recommendation framework, memory-based $\textbf{P}$ersonalized $\textbf{R}$ecommendation $\textbf{T}$ool learning via autonomous language $\textbf{A}$gents (PRTA), in which an LLM acts as a central planner interacting with multiple recommendation models as tools. The LLM-based agent is responsible for high-level reasoning and personalized tool selection, while traditional recommendation models perform full-ranking scoring, leveraging their scalability in modeling behavioral patterns. To support personalized tool selection, we design reflection mechanisms that enable the agent to evaluate and compare tools for each user based on user profiles and candidate ranked lists. Extensive experiments across three public datasets demonstrate the superiority of \modelname over traditional recommendation and LLM-based baselines in improving full-ranking recommendation performance.

Problem

• personalized recommendation
• large language models
• full-ranking
• hallucination
• context-length limitation

Innovation

• autonomous language agents
• tool learning
• personalized recommendation
• large language models
• full-ranking recommendation