Quantifying Privacy Risks of Masked Language Models Using Membership Inference Attacks

Fatemehsadat Mireshghallah, Kartik Goyal, Archit Uniyal, Taylor Berg-Kirkpatrick, Reza Shokri