ARAG: Agentic Retrieval Augmented Generation for Personalized Recommendation

TL;DR

ARAG introduces a multi-agent framework that enhances personalized recommendations by integrating user understanding, semantic evaluation, and ranking agents into retrieval-augmented generation, significantly outperforming baseline methods.

Contribution

This work presents ARAG, a novel multi-agent RAG framework that captures nuanced user preferences and improves recommendation quality through agentic reasoning.

Findings

ARAG outperforms standard RAG and recency baselines in NDCG@5 and Hit@5.

Experimental results show up to 42.1% improvement in NDCG@5.

Ablation study confirms the effectiveness of each ARAG component.

Abstract

Retrieval-Augmented Generation (RAG) has shown promise in enhancing recommendation systems by incorporating external context into large language model prompts. However, existing RAG-based approaches often rely on static retrieval heuristics and fail to capture nuanced user preferences in dynamic recommendation scenarios. In this work, we introduce ARAG, an Agentic Retrieval-Augmented Generation framework for Personalized Recommendation, which integrates a multi-agent collaboration mechanism into the RAG pipeline. To better understand the long-term and session behavior of the user, ARAG leverages four specialized LLM-based agents: a User Understanding Agent that summarizes user preferences from long-term and session contexts, a Natural Language Inference (NLI) Agent that evaluates semantic alignment between candidate items retrieved by RAG and inferred intent, a context summary agent that summarizes the findings of NLI agent, and an Item Ranker Agent that generates a ranked list of recommendations based on contextual fit. We evaluate ARAG accross three datasets. Experimental results demonstrate that ARAG significantly outperforms standard RAG and recency-based baselines, achieving up to 42.1% improvement in NDCG@5 and 35.5% in Hit@5. We also, conduct an ablation study to analyse the effect by different components of ARAG. Our findings highlight the effectiveness of integrating agentic reasoning into retrieval-augmented recommendation and provide new directions for LLM-based personalization.