LLM-based user agents, which simulate user interaction behavior, are emerging as a promising approach to enhancing recommender systems. In real-world scenarios, users' interactions often exhibit cross-domain characteristics and are influenced by others. However, the memory design in current methods causes user agents to introduce significant irrelevant information during decision-making in cross-domain scenarios and makes them unable to recognize the influence of other users' interactions, such as popularity factors. To tackle this issue, we propose a dual-layer memory architecture combined with a two-step fusion mechanism. This design avoids irrelevant information during decision-making while ensuring effective integration of cross-domain preferences. We also introduce the concepts of interest groups and group-shared memory to better capture the influence of popularity factors on users with similar interests. Comprehensive experiments validate the effectiveness of AgentCF++. Our code is available at https://github.com/jhliu0807/AgentCF-plus.
@article{arxiv.2502.13843,
title = {AgentCF++: Memory-enhanced LLM-based Agents for Popularity-aware Cross-domain Recommendations},
author = {Jiahao Liu and Shengkang Gu and Dongsheng Li and Guangping Zhang and Mingzhe Han and Hansu Gu and Peng Zhang and Tun Lu and Li Shang and Ning Gu},
journal= {arXiv preprint arXiv:2502.13843},
year = {2025}
}