FFHow

Model-Inversion-Attack-ToolBox. A comprehensive toolbox for model inversion attacks and defenses, which is easy to get started.

GIFD_Gradient_Inversion_Attack. [ICCV-2023] Gradient inversion attack, Federated learning, Generative adversarial network.

CGNC_Targeted_Adversarial_Attacks. [ECCV-2024] Transferable Targeted Adversarial Attack, CLIP models, Generative adversarial network, Multi-target attacks

CPGC_VLP_Universal_Attacks. [ICCV-2025] Universal Adversarial Attack, Multimodal Adversarial Attacks, VLP models, Contrastive Learning, Cross-modal Perturbation Generator, Generative Attacks.

CoPA_Contrastive_Paraphrase_Attacks. [EMNLP-2025] Paraphrasing Attacks, LLM-Generated Text Detectors, Training-free Decoding Strategy

CMI_VLD_Hallucination_Mitigation. [NeurIPS-2025] Large Vision-Language Models, Hallucination Mitigation, Conditional Mutual Information, Token Purification

BadRDM_Backdoor_RAG_diffusion_models. [ACL-2026] Backdoor attack, Retrieval Augmented Diffusion Models.

Visualization_Code. Python code for ploting figures, though a little bit messy.