
This paper introduces Lastde, a novel training-free method for detecting LLM-generated text by analyzing the temporal dynamics of token probability sequences using diversity entropy, achieving state-of-the-art performance in both white-box and black-box settings.


coremsg

training-free-detection-of-llm-generated-text-using-token-probability-sequence-analysis-


Training-Free Detection of LLM-Generated Text Using Token Probability Sequence Analysis


title_rewrite


Detecting text generated by large language models (LLMs) is crucial to mitigate potential misuse and safeguard realms like artistic expression and social networks from harmful influence of LLM-generated content.


comprehensive-survey-on-detecting-text-generated-by-large-language-models-necessity-methods-and-future-directions


Comprehensive Survey on Detecting Text Generated by Large Language Models: Necessity, Methods, and Future Directions
