# Safeguarding Multimodal Knowledge Copyright in the RAG-as-a-Service Environment Source: [Safeguarding Multimodal Knowledge Copyright in the RAG-as-a-Service Environment]() Collection index: [llms.txt]() Data updated: 2026-03-04T05:00:05.078Z Authors: Tianyu Chen, Jian Lou, Wenjie Wang Published: 2025-06-10T09:56:02.000Z ## Abstract As Retrieval-Augmented Generation (RAG) evolves into service-oriented platforms (Rag-as-a-Service) with shared knowledge bases, protecting the copyright of contributed data becomes essential. Existing watermarking methods in RAG focus solely on textual knowledge, leaving image knowledge unprotected. In this work, we propose AQUA, the first watermark framework for image knowledge protection in Multimodal RAG systems. AQUA embeds semantic signals into synthetic images using two complementary methods: acronym-based triggers and spatial relationship cues. These techniques ensure watermark signals survive indirect watermark propagation from image retriever to textual generator, being efficient, effective and imperceptible. Experiments across diverse models and datasets show that AQUA enables robust, stealthy, and reliable copyright tracing, filling a key gap in multimodal RAG protection. ## Recorded benchmark results A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. - [Watermark Detection on MMQA](): Watermark Detection; MMQA; Rank: 1.03. - [Watermark Detection on WebQA](): Watermark Detection; WebQA; Rank: 1.05. ## Original sources and implementation [Original paper]() [PDF]()