An efficient watermarking method for latent diffusion models via low-rank adaptation and dynamic loss weighting
- 5 June 2026
- journal article
- Published by Elsevier in Expert Systems with Applications
- Vol. 331
Abstract
No abstract availableAll Related Versions
Funding Information
- Natural Science Foundation of Xiamen Municipality (3502Z202572015)
- European Commission
- Natural Science Foundation of Fujian Province (2026J001762)
- Shenzhen Research and Development Program
- National Natural Science Foundation of China (U22B2047)
- National Natural Science Foundation of China (U23B2022)
This publication has 17 references indexed in Scilit:
- Towards provenance-aware diffusion: Key-free watermarking with Gaussian shadingExpert Systems with Applications, 2025
- Distributor-centric model watermarking for image generative modelsKnowledge-Based Systems, 2025
- Model authentication hashing: Identifying pirated neural network modelsExpert Systems with Applications, 2025
- Intellectual Property Protection of Diffusion Models via the Watermark Diffusion Process2024
- Watermarking for Stable Diffusion ModelsIEEE Internet of Things Journal, 2024
- A novel robust black-box fingerprinting scheme for deep classification neural networksExpert Systems with Applications, 2024
- Universal BlackMarks: Key-Image-Free Blackbox Multi-Bit Watermarking of Deep Neural NetworksIEEE Signal Processing Letters, 2023
- Intellectual Property Protection for Deep Learning Models: Taxonomy, Methods, Attacks, and EvaluationsIEEE Transactions on Artificial Intelligence, 2021
- A survey of Deep Neural Network watermarking techniquesNeurocomputing, 2021
- ReDMark: Framework for residual diffusion watermarking based on deep networksExpert Systems with Applications, 2020