Risk Management in Image Generative Models Through Model Fingerprinting
Abstract
My doctoral research delves into the realm of generative model fingerprinting, aiming to assign responsibility for the generated images. I introduce frameworks that modify generative models to incorporate each user's distinct digital fingerprint. This ensures that every piece of generated content carries a traceable identifier linked to its originator. The primary objective of my research is to achieve optimal attribution accuracy while ensuring minimal compromise on the model's performance. Additionally, I present strategies designed to enhance robustness against common adversarial manipulations, which malicious users might employ to obscure or remove these fingerprints.
Cite
Text
Kim. "Risk Management in Image Generative Models Through Model Fingerprinting." AAAI Conference on Artificial Intelligence, 2024. doi:10.1609/AAAI.V38I21.30397Markdown
[Kim. "Risk Management in Image Generative Models Through Model Fingerprinting." AAAI Conference on Artificial Intelligence, 2024.](https://mlanthology.org/aaai/2024/kim2024aaai-risk/) doi:10.1609/AAAI.V38I21.30397BibTeX
@inproceedings{kim2024aaai-risk,
title = {{Risk Management in Image Generative Models Through Model Fingerprinting}},
author = {Kim, Changhoon},
booktitle = {AAAI Conference on Artificial Intelligence},
year = {2024},
pages = {23393-23394},
doi = {10.1609/AAAI.V38I21.30397},
url = {https://mlanthology.org/aaai/2024/kim2024aaai-risk/}
}