揭示大模型推荐系统中风险的递归演化机制
EchoTrace: Diagnosing Recursive Risks in LLM-Powered Recommender Systems
- 分阶段诊断框架识别大模型在推荐中的风险生成路径
- 实验证明大模型会放大流行度偏差并产生虚假信号
- 适合关注大模型推荐系统安全性的研究人员和工程师
大型语言模型（LLMs）正被广泛用于推荐系统中作为数据增强、用户画像生成和推荐模块。尽管这些角色能提升语义理解与推荐质量，但也引入了偏见和幻觉等特定风险。在反馈循环场景下，这些风险会随时间递归传播，影响训练数据和推荐动态。本文提出一种角色感知、分阶段的诊断框架，通过受控反馈循环模拟与纵向分析，追踪风险在大模型生成内容、推荐输出、反馈循环及生态层面的演进过程。在多个主流基准测试中，实验表明大模型组件会加剧流行度偏差，通过幻觉引入虚假信号，并随时间逐步形成极化且自我强化的曝光模式。代码已开源。
原文摘要 · Abstract (English)
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.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。