Zi Liang

Privacy (1)

large language model (3) model extraction (2) reinforcement learning (1) benchmark evaluation (1) semi-supervised learning (1) policy gradient (1) preference learning (1) knowledge distillation (1) text generation (1) point cloud (1) cross-modal retrieval (1) machine unlearning (1) intellectual property (1) safety alignment (1) evaluation framework (1) reinforcement learning from human feedback (1) privacy preserving (1) language model (1) 3d representation (1) membership inference (1)