ownership-verification

Proposes a unified framework, LCF and LCFEdit, that jointly optimizes construction and injection of code-mixing fingerprints for LLMs using low-resource languages to achieve imperceptible and robust ownership verification.

FedOT introduces a chunked watermarking and latent vector transformation framework for ownership verification and leakage tracing in federated latent diffusion models, preventing watermark removal attacks.

This paper introduces CopyCop, an algorithm for verifying ownership of Graph Neural Networks by detecting surrogate models even when they differ in architecture, weights, or output transformations.