循环中的回声:在LLM推荐系统中反馈循环下的风险诊断
EchoTrace: Diagnosing Recursive Risks in LLM-Powered Recommender Systems
AI总结:
本文提出了一种角色感知的诊断框架,用于分析LLM推荐系统中反馈循环下的系统性风险,包括偏见、幻觉和自我强化的暴露模式。
摘要:
大型语言模型(LLMs)越来越多地嵌入到推荐系统中,其中它们在多个功能角色上发挥作用,例如数据增强、用户画像和决策制定。尽管先前的研究强调推荐性能,但LLM的系统性风险,如偏见和幻觉,以及它们通过反馈循环传播的风险仍鲜有研究。在本文中,我们提出了一种角色感知、分阶段的诊断框架,追踪这些风险如何产生、在排序结果中体现,并在重复推荐循环中累积。我们正式化了一个受控的反馈循环管道,模拟长期的交互动态,并在LLM生成内容、排序和生态系统层面实现经验测量。实验结果表明,基于LLM的组件可以放大流行偏见,通过幻觉引入虚假信号,并导致随时间推移而极化和自我强化的暴露模式。我们计划将我们的框架作为开源工具包发布,以促进在多样化的LLM推荐系统中的系统性风险分析。
英文摘要:
Large language models (LLMs) are increasingly integrated into recommender systems as data augmenters, profile generators, and recommendation modules. While these roles can enhance semantic understanding and recommendation quality, they also introduce LLM-specific risks such as bias and hallucination. These risks become more critical in feedback-loop settings, where LLM-generated signals and recommendations recursively shape future training data and recommendation dynamics. In this paper, we propose a role-aware, phase-wise diagnostic framework for analyzing how LLM-induced risks emerge, propagate, and accumulate in LLM-powered recommender systems. Our framework combines controlled feedback-loop simulation with longitudinal phase-wise diagnosis across LLM-generated content, recommendation outputs, feedback-loop dynamics, and ecosystem-level effects. Experiments on widely used benchmarks show that LLM-based components can amplify popularity bias, introduce spurious signals through hallucination, and gradually produce polarized and self-reinforcing exposure patterns over time. The code for EchoTrace is available at https://github.com/DongUk-Park/EchoTrace.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。