Fetching the paper…

Watermarking Vision-Language Pre-trained Models for Multi-modal Embedding as a Service · Around