LLM as a Broken Telephone: Iterative Generation Distorts Information
Mohamed, Amr Geng, Mingmeng Vazirgiannis, Michalis Shang, Guokan
Mohamed, Amr
Geng, Mingmeng
Vazirgiannis, Michalis
Shang, Guokan
Supervisor
Department
Machine Learning
Embargo End Date
Type
Conference proceeding
Date
2025
License
Language
English
Collections
Research Projects
Organizational Units
Journal Issue
Abstract
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.
Citation
A. Mohamed, M. Geng, M. Vazirgiannis, G. Shang, and E. Polytechnique, “LLM as a Broken Telephone: Iterative Generation Distorts Information,” vol. 1, pp. 7493–7509, Aug. 2025, doi: 10.18653/V1/2025.ACL-LONG.371.
Source
Proceedings of the Annual Meeting of the Association for Computational Linguistics
Conference
63rd Annual Meeting of the Association for Computational Linguistics, ACL 2025
Keywords
Iterative Generation, Large Language Models, Information Distortion, Chained Generation, Translation-based Experimentation, Model Output Accumulation, Prompting Mitigation Strategies, Reliability of LLM-Generated Content
Subjects
Source
63rd Annual Meeting of the Association for Computational Linguistics, ACL 2025
Publisher
Association for Computational Linguistics