Hey! a RS from Meta seamless team here.

Yes, we chose not to release the watermark detector to safeguard against adversarial attacks. This decision helps prevent any attempts to erase the watermark by malicious users.

The watermark generator and detector are trained together, one can use the information in our paper to train your own generator and detector model, however in this case the watermark signature created will be distinct from the one we use to protect our seamless translation models. This approach ensures each model maintains its unique security features.

Thanks for clarifying, and seems like a completely reasonable approach. Thanks for the great work.