Yujian Sun

large language model (3) direct preference optimization (2) multi-label classification (2) emotion detection (2) selective forgetting (2) contrastive learning (2) machine unlearning (2) emotion intensity prediction (2) knowledge erasure (2) sensitive content (1) multi-task framework (1) llm unlearning (1) contrastive reasoning calibration (1) model preservation (1) model compression (1) unlearning loss (1) multi-task learning (1) responsible ai (1) model editing (1) instruction tuning (1)