Amber Yijia Zheng

IMMA. [ECCV2024] Immunizing text-to-image Models against Malicious Adaptation

vae-cvae-manifold-dim. [NeurIPS 2022] Learning Manifold Dimensions with Conditional Variational Autoencoders

model-immunization-cond-num. [ICML 2025 Oral] Model Immunization from a Condition Number Perspective

MIMA. [AAAI 2025] Multi-concept Model Immunization through Differentiable Model Merging

spm_unlearning. [CVPR 2026 Highlight] Designing to Forget: Deep Semi-parametric Models for Unlearning

LTO. [ECCV2024] Learning to Obstruct Few-Shot Image Classification over Restricted Classes

BloomGML. Code for Graph Machine Learning through the Lens of Bilevel Optimization