DF-MIA: A Distribution-Free Membership Inference Attack on Fine-Tuned Large Language Models.

Zhiheng Huang, Yannan Liu, Daojing He, Yu Li

Browse the full AAAI paper archive.