Generative AI model privacy: a survey

Artificial Intelligence Review, Journal Year: 2024, Volume and Issue: 58(1)

Published: Dec. 4, 2024

Abstract The rapid progress of generative AI models has yielded substantial breakthroughs in AI, facilitating the generation realistic synthetic data across various modalities. However, these advancements also introduce significant privacy risks, as may inadvertently expose sensitive information from their training data. Currently, there is no comprehensive survey work investigating issues, e.g., attacking and defending models. We strive to identify existing attack techniques mitigation strategies offer a summary current research landscape. Our encompasses wide array models, including language Generative Adversarial Networks, diffusion multi-modal counterparts. It indicates critical need for continued development privacy-preserving Furthermore, we insights into challenges discuss open problems intersection

Language: Английский

---

ChatGPT Knows Your Attacks: Synthesizing Attack Trees Using LLMs

Communications in computer and information science, Journal Year: 2023, Volume and Issue: unknown, P. 245 - 260

Published: Jan. 1, 2023

Language: Английский

---

What, Indeed, is an Achievable Provable Guarantee for Learning-Enabled Safety-Critical Systems

Lecture notes in computer science, Journal Year: 2023, Volume and Issue: unknown, P. 55 - 76

Published: Dec. 13, 2023

Language: Английский

---

LUNA: A Model-Based Universal Analysis Framework for Large Language Models

IEEE Transactions on Software Engineering, Journal Year: 2024, Volume and Issue: 50(7), P. 1921 - 1948

Published: July 1, 2024

---

Towards Fairness-Aware Adversarial Learning

2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Journal Year: 2024, Volume and Issue: 33, P. 24746 - 24755

Published: June 16, 2024

Language: Английский

---