Large language models (LLMs) are increasingly embedded into recommender systems, where they operate across multiple functional roles such as data augmentation, profiling, and decision making. While prior work emphasizes recommendation performance, the systemic risks of LLMs, such as bias and hallucination, and their propagation through feedback loops remain largely unexplored. In this paper, we propose a role-aware, phase-wise diagnostic framework that traces how these risks emerge, manifest in ranking outcomes, and accumulate over repeated recommendation cycles. We formalize a controlled feedback-loop pipeline that simulates long-term interaction dynamics and enables empirical measurement of risks at the LLM-generated content, ranking, and ecosystem levels. Experiments on widely used benchmarks demonstrate that LLM-based components can amplify popularity bias, introduce spurious signals through hallucination, and lead to polarized and self-reinforcing exposure patterns over time. We plan to release our framework as an open-source toolkit to facilitate systematic risk analysis across diverse LLM-powered recommender systems.

翻译：大型语言模型（LLM）正日益嵌入推荐系统中，承担数据增强、用户画像构建及决策制定等多种功能角色。尽管已有研究主要关注推荐性能，但LLM的系统性风险（如偏见与幻觉）及其通过反馈循环的传播机制仍鲜有探讨。本文提出一种角色感知、分阶段诊断框架，用以追踪这些风险如何产生、在排序结果中显现，并在重复推荐周期中不断累积。我们形式化了一个受控反馈循环流程，该流程模拟长期交互动态，并支持在LLM生成内容、排序及生态系统三个层面对风险进行实证度量。在广泛使用的基准测试上的实验表明，基于LLM的组件可能放大流行度偏见、通过幻觉引入虚假信号，并随时间推移导致极化且自我强化的曝光模式。我们计划将本框架作为开源工具包发布，以促进对不同类型基于LLM的推荐系统进行系统性风险分析。