Qizhang Feng

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

Gradient Rewiring for Editable Graph Neural Network Training

UNITE:Universally Trustworthy GNN Via Subgraph Identification