AMEM4Rec: Leveraging Cross-User Similarity for Memory Evolution in Agentic LLM Recommenders

TL;DR

AMEM4Rec introduces a novel agentic LLM recommender that learns collaborative filtering signals through cross-user memory evolution, improving recommendation accuracy without relying on pre-trained CF models.

Contribution

It proposes an end-to-end memory evolution approach that captures collaborative signals in LLM recommenders, addressing limitations of fine-tuning and prompt-based methods.

Findings

Outperforms state-of-the-art LLM recommenders on Amazon and MIND datasets.

Effectively captures collaborative filtering signals without pre-trained CF models.

Demonstrates the benefit of memory-guided cross-user pattern evolution.

Abstract

Agentic systems powered by Large Language Models (LLMs) have shown strong potential in recommender systems but remain hindered by several challenges. Fine-tuning LLMs is parameter-inefficient, and prompt-based agentic reasoning is limited by context length and hallucination risk. Moreover, existing agentic recommendation systems predominantly leverages semantic knowledge while neglecting the collaborative filtering (CF) signals essential for implicit preference modeling. To address these limitations, we propose AMEM4Rec, an agentic LLM-based recommender that learns collaborative signals in an end-to-end manner through cross-user memory evolution. AMEM4Rec stores abstract user behavior patterns from user histories in a global memory pool. Within this pool, memories are linked to similar existing ones and iteratively evolved to reinforce shared cross-user patterns, enabling the system to become aware of CF signals without relying on a pre-trained CF model. Extensive experiments on Amazon and MIND datasets show that AMEM4Rec consistently outperforms state-of-the-art LLM-based recommenders, demonstrating the effectiveness of evolving memory-guided collaborative filtering.