Michalis Vazirgiannis, Guokan Shang, Mingmeng Geng, Amr Mohamed
We have not lifted any functions out of this paper's repositories yet, so there is nothing we have run. If it links a repository, it is listed below.
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 AImediated information propagation, raising important questions about the reliability of LLMgenerated content in iterative workflows.
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("2502.20258")
get_code_for_paper("2502.20258")
have("2502.20258")
Connect an agent — have() is free.