WARDEN: Multi-Directional Backdoor Watermarks for Embedding-as-a-Service Copyright Protection

arXiv:2403.01472v1 Announce Type: new
Abstract: Embedding as a Service (EaaS) has become a widely adopted solution, which offers feature extraction capabilities for addressing various downstream tasks in Natural Language Processing (NLP). Prior studies have shown that EaaS can be prone to model extraction attacks; nevertheless, this concern could be mitigated by adding backdoor watermarks to the text embeddings and subsequently verifying the attack models post-publication. Through the analysis of the recent watermarking strategy for EaaS, EmbMarker, we design a novel …

arxiv as-a-service attacks backdoor can capabilities copyright copyright protection cs.cl cs.cr cs.lg eaas extraction feature language model extraction natural natural language natural language processing nlp protection service solution studies watermarks