Large Language Model Turnover Undermines Screening for Artificial Intelligence-Assisted Scientific Writing.

arXiv:2610.11599v1 Announce Type: cross Abstract: Journals and conferences have begun to screen submitted manuscripts for text written using large language models (LLMs). The reliability of this screening rests on benchmark evaluations against a fixed set of L