Siva Rajesh Kasa

The Hidden Cost of Modeling $\text{P}(X)$: Membership Inference Attacks in Generative Text Classifiers

Exposing Privacy Gaps: Membership Inference Attack on Preference Data for LLM Alignment