watermarking
Watermarking in AI refers to embedding unique identifiers into models or datasets to indicate ownership or to track usage, which aids in preventing unauthorized use or replicating proprietary information, particularly critical in commercial applications.
- BitMark: Watermarking Bitwise Autoregressive Image Generative Models
- Enhancing LLM Watermark Resilience Against Both Scrubbing and Spoofing Attacks
- Mask Image Watermarking
- On the Coexistence and Ensembling of Watermarks
- Robust Distortion-Free Watermark for Autoregressive Audio Generation Models
- SAEMark: Steering Personalized Multilingual LLM Watermarks with Sparse Autoencoders
- Theoretically Grounded Framework for LLM Watermarking: A Distribution-Adaptive Approach
- Watermarking Autoregressive Image Generation