Anh Tuan Bui

Antibody: Strengthening Defense Against Harmful Fine-Tuning for Large Language Models via Attenuating Harmful Gradient Influence

Mitigating Semantic Collapse in Generative Personalization with Test-Time Embedding Adjustment

A Concept is More Than a Word: Diversified Unlearning in Text-to-Image Diffusion Models

Improving LLM Unlearning Robustness via Random Perturbations

Fantastic Targets for Concept Erasure in Diffusion Models and Where To Find Them

Unified Perspectives on Signal-to-Noise Diffusion Models

Unified Deep Discrete Representation Learning Framework