mlsanityData to 7 Oct 2026

RegionMarker: A Region-Triggered Semantic Watermarking Framework for Embedding-as-a-Service Copyright Protection

arXiv 2511.13329 ↗First posted 17 Nov 2025cs.CL cs.CRcited by 4

Not on the map. Why?

Abstract (arXiv)

Embedding-as-a-Service (EaaS) is an effective and convenient deployment solution for addressing various NLP tasks. Nevertheless, recent research has shown that EaaS is vulnerable to model extraction attacks, which could lead to significant economic losses for model providers. For copyright protection, existing methods inject watermark embeddings into text embeddings and use them to detect copyright infringement. However, current watermarking methods often resist only a subset of attacks and fail to provide textitcomprehensive protection. To this end, we present the region-triggered semantic watermarking framework called RegionMarker, which defines trigger regions within a low-dimensional space and injects watermarks into text embeddings associated with these regions. By utilizing a secret dimensionality reduction matrix to project onto this subspace and randomly selecting trigger regions, RegionMarker makes it difficult for watermark removal attacks to evade detection. Furthermore, by embedding watermarks across the entire trigger region and using the text embedding as the watermark, RegionMarker is resilient to both paraphrasing and dimension-perturbation attacks. Extensive experiments on various datasets show that RegionMarker is effective in resisting different attack methods, thereby protecting the copyright of EaaS.

Key facts. Cited by 4 papers in the map (data through 7 Oct 2026). Builds on 23 of its 37 references that are in the map. First posted on arXiv 17 Nov 2025.

Builds on

23 of 37 references.

References · 23

  1. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,204
  2. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 13,958
  3. 2015Character-level Convolutional Networks for Text Classificationcited by 2,293
  4. 2022No Language Left Behind: Scaling Human-Centered Machine Translationcited by 1,290
  5. 2016Stealing Machine Learning Models via Prediction APIscited by 602
  6. 2024NV-Embed: Improved Techniques for Training LLMs as Generalist Embedding Modelscited by 466
  7. 2023Paraphrasing evades detectors of AI-generated text, but retrieval is an effective defensecited by 401
  8. 2018Knockoff Nets: Stealing Functionality of Black-Box Modelscited by 285
  9. 2019Thieves on Sesame Street! Model Extraction of BERT-based APIscited by 136
  10. 2024V-Express: Conditional Dropout for Progressive Training of Portrait Video Generationcited by 85
  11. 2018Exploring Connections Between Active Learning and Model Extractioncited by 84
  12. 2020Imitation Attacks and Defenses for Black-box Machine Translation Systemscited by 78
11 more
  1. 2023Are You Copying My Model? Protecting the Copyright of Large Language Models for EaaS via Backdoor Watermarkcited by 61
  2. 2023BeLLM: Backward Dependency Enhanced Large Language Model for Sentence Embeddingscited by 32
  3. 2024Token Prepending: A Training-Free Approach for Eliciting Better Sentence Embeddings from LLMscited by 18
  4. 2025Contrastive Prompting Enhances Sentence Embeddings in LLMs through Inference-Time Steeringcited by 13
  5. 2024Ensembling Diffusion Models via Adaptive Feature Aggregationcited by 12
  6. 2025UORA: Uniform Orthogonal Reinitialization Adaptation in Parameter-Efficient Fine-Tuning of Large Modelscited by 11
  7. 2024WARDEN: Multi-Directional Backdoor Watermarks for Embedding-as-a-Service Copyright Protectioncited by 9
  8. 2025Advanced Sign Language Video Generation with Compressed and Quantized Multi-Condition Tokenizationcited by 9
  9. 2025Pretraining on the Test Set Is No Longer All You Need: A Debate-Driven Approach to QA Benchmarkscited by 7
  10. 2025Steering When Necessary: Flexible Steering Large Language Models with Backtrackingcited by 7
  11. 2024WET: Overcoming Paraphrasing Vulnerabilities in Embeddings-as-a-Service with Linear Transformation Watermarkscited by 5

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 4.

4 papers.

  1. 2026SlowBA: An efficiency backdoor attack towards VLM-based GUI agentscited by 3
  2. 2026textbfAGTAO: Robust and Stabilized LLM Unlearning via Adversarial Gating Training with Adaptive Orthogonalitycited by 0
  3. 2026Geometry-Aware Localized Watermarking for Copyright Protection in Embedding-as-a-Servicecited by 0
  4. 2026Focusing Condition: Inference-Time Self-Contrastive Steering Elicits Better Conditional Text Embeddings in LLMscited by 0