mlsanityData to 14 Oct 2026

LLM as a Broken Telephone: Iterative Generation Distorts Information

arXiv 2502.20258 ↗First posted 27 Feb 2025cs.CL cs.AIcited by 12

Not on the map. Why?

Abstract (arXiv)

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.

Key facts. Cited by 12 papers in the map (data through 14 Oct 2026). Builds on 27 of its 46 references that are in the map. First posted on arXiv 27 Feb 2025.

Builds on

27 of 46 references.

References · 27

  1. 2024The Llama 3 Herd of Modelscited by 14,715
  2. 2019BERTScore: Evaluating Text Generation with BERTcited by 5,512
  3. 2023Generative Agents: Interactive Simulacra of Human Behaviorcited by 3,568
  4. 2024Gemma 2: Improving Open Language Models at a Practical Sizecited by 1,998
  5. 2023AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversationcited by 1,988
  6. 2023FActScore: Fine-grained Atomic Evaluation of Factual Precision in Long Form Text Generationcited by 1,282
  7. 2020BLEURT: Learning Robust Metrics for Text Generationcited by 1,203
  8. 2024Scaling Synthetic Data Creation with 1,000,000,000 Personascited by 417
  9. 2023The Curse of Recursion: Training on Generated Data Makes Models Forgetcited by 356
  10. 2023FacTool: Factuality Detection in Generative AI -- A Tool Augmented Framework for Multi-Task and Multi-Domain Scenarioscited by 251
  11. 2023Exploring Collaboration Mechanisms for LLM Agents: A Social Psychology Viewcited by 222
  12. 2021BookSum: A Collection of Datasets for Long-form Narrative Summarizationcited by 184
15 more
  1. 2024A Tale of Tails: Model Collapse as a Change of Scaling Lawscited by 130
  2. 2024Improving Multi-Agent Debate with Sparse Communication Topologycited by 107
  3. 2023The Curious Decline of Linguistic Diversity: Training Language Models on Synthetic Textcited by 100
  4. 2023LatestEval: Addressing Data Contamination in Language Model Evaluation through Dynamic and Time-Sensitive Test Constructioncited by 58
  5. 2023Let Models Speak Ciphers: Multiagent Debate through Embeddingscited by 51
  6. 2024Standardizing the Measurement of Text Diversity: A Tool and a Comparative Analysis of Scorescited by 50
  7. 2020When Does Unsupervised Machine Translation Work?cited by 38
  8. 2024AI and the Problem of Knowledge Collapsecited by 34
  9. 2024Towards Theoretical Understandings of Self-Consuming Generative Modelscited by 29
  10. 2024OpenFactCheck: A Unified Framework for Factuality Evaluation of LLMscited by 28
  11. 2023Does Human Collaboration Enhance the Accuracy of Identifying LLM-Generated Deepfake Texts?cited by 21
  12. 2022Evaluating Distributional Distortion in Neural Language Modelingcited by 19
  13. 2024Tracking the perspectives of interacting language modelscited by 14
  14. 2024To Diverge or Not to Diverge: A Morphosyntactic Perspective on Machine Translation vs Human Translationcited by 10
  15. 2024The Impact of Syntactic and Semantic Proximity on Machine Translation with Back-Translationcited by 2

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 3; 2026: 9.

12 papers.

  1. 2025Lost in the Mix: Evaluating LLM Understanding of Code-Switched Textcited by 10
  2. 2025Wikipedia in the Era of LLMs: Evolution and Riskscited by 4
  3. 2026Mutation Without Variation: Convergence Dynamics in LLM-Driven Program Evolutioncited by 4
  4. 2025Another Turn, Better Output? A Turn-Wise Analysis of Iterative LLM Promptingcited by 3
  5. 2026Markovian Generation Chains in Large Language Modelscited by 3
  6. 2026When Summaries Distort Decisions: Information Fidelity in LLM-Compressed Financial Analysiscited by 2
  7. 2026Beyond Via: Analysis and Estimation of the Impact of Large Language Models in Academic Paperscited by 1
  8. 2026HawkesLLM: Semantic Uncertainty Propagation in Agentic Text Simulationcited by 0
  9. 2026From 'May' to 'Is': Certainty Distortion in Language Model Rewritingcited by 0
  10. 2026Faithful, Not Corrective: Model Capability Governs Message-Format Effects in Multi-Hop Agent Relayscited by 0
  11. 2026Long-Horizon AI Research for Grothendieck Constant: A Case Study in Human-AI Mathematical Collaborationcited by 0
  12. 2026If It's Not Buggy, Don't Fix It: On the Dynamics of Iterative Bug-fixing with LLMscited by 0