大语言模型如同失真的传声筒：迭代生成导致信息失真

LLM as a Broken Telephone: Iterative Generation Distorts Information

February 27, 2025

作者: Amr Mohamed, Mingmeng Geng, Michalis Vazirgiannis, Guokan Shang

cs.AI

摘要

随着大型语言模型日益承担起在线内容生成的责任，人们开始担忧其反复处理自身输出所产生的影响。受人类链式沟通中"传话失真"现象的启发，本研究探讨了LLM是否也会在迭代生成过程中类似地扭曲信息。通过基于翻译的实验，我们发现失真会随时间累积，并受到语言选择和链条复杂性的影响。虽然质量下降不可避免，但通过策略性的提示技术可以缓解这一问题。这些发现为讨论AI中介信息传播的长期效应提供了依据，并引发了关于迭代工作流程中LLM生成内容可靠性的重要问题。

English

As large language models are increasingly responsible for online content, concerns arise about the impact of repeatedly processing their own outputs. Inspired by the "broken telephone" effect in chained human communication, this study investigates whether LLMs similarly distort information through iterative generation. Through translation-based experiments, we find that distortion accumulates over time, influenced by language choice and chain complexity. While degradation is inevitable, it can be mitigated through strategic prompting techniques. These findings contribute to discussions on the long-term effects of AI-mediated information propagation, raising important questions about the reliability of LLM-generated content in iterative workflows.