mlsanityData to 7 Oct 2026

WET: Overcoming Paraphrasing Vulnerabilities in Embeddings-as-a-Service with Linear Transformation Watermarks

arXiv 2409.04459 ↗First posted 29 Aug 2024cs.CR cs.CL cs.LGcited by 5

Not on the map. Why?

Abstract (arXiv)

Embeddings-as-a-Service (EaaS) is a service offered by large language model (LLM) developers to supply embeddings generated by LLMs. Previous research suggests that EaaS is prone to imitation attacks -- attacks that clone the underlying EaaS model by training another model on the queried embeddings. As a result, EaaS watermarks are introduced to protect the intellectual property of EaaS providers. In this paper, we first show that existing EaaS watermarks can be removed by paraphrasing when attackers clone the model. Subsequently, we propose a novel watermarking technique that involves linearly transforming the embeddings, and show that it is empirically and theoretically robust against paraphrasing.

Key facts. Cited by 5 papers in the map (data through 7 Oct 2026). Builds on 21 of its 37 references that are in the map. First posted on arXiv 29 Aug 2024.

Builds on

21 of 37 references.

References · 21

  1. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,204
  2. 2015Character-level Convolutional Networks for Text Classificationcited by 2,293
  3. 2022No Language Left Behind: Scaling Human-Centered Machine Translationcited by 1,290
  4. 2023DetectGPT: Zero-Shot Machine-Generated Text Detection using Probability Curvaturecited by 705
  5. 2016Stealing Machine Learning Models via Prediction APIscited by 602
  6. 2021Datasets: A Community Library for Natural Language Processingcited by 526
  7. 2023Paraphrasing evades detectors of AI-generated text, but retrieval is an effective defensecited by 401
  8. 2023Can AI-Generated Text be Reliably Detected?cited by 372
  9. 2018Knockoff Nets: Stealing Functionality of Black-Box Modelscited by 285
  10. 2023Robust Distortion-free Watermarks for Language Modelscited by 279
  11. 2023Text Embeddings Reveal (Almost) As Much As Textcited by 189
  12. 2019Thieves on Sesame Street! Model Extraction of BERT-based APIscited by 136
9 more
  1. 2020Imitation Attacks and Defenses for Black-box Machine Translation Systemscited by 78
  2. 2021Protecting Intellectual Property of Language Generation APIs with Lexical Watermarkcited by 78
  3. 2022CATER: Intellectual Property Protection on Text Generation APIs via Conditional Watermarkscited by 68
  4. 2023Are You Copying My Model? Protecting the Copyright of Large Language Models for EaaS via Backdoor Watermarkcited by 61
  5. 2021Model Extraction and Adversarial Transferability, Your BERT is Vulnerable!cited by 51
  6. 2024Text Embedding Inversion Security for Multilingual Language Modelscited by 26
  7. 2023Rethinking White-Box Watermarks on Deep Learning Models under Neural Structural Obfuscationcited by 20
  8. 2021Student Surpasses Teacher: Imitation Attack for Black-Box NLP APIscited by 16
  9. 2024WARDEN: Multi-Directional Backdoor Watermarks for Embedding-as-a-Service Copyright Protectioncited by 9

Built on by

Cited by year

Citing papers per year: 2024: 3; 2025: 2.

5 papers.

  1. 2024Robust and Minimally Invasive Watermarking for EaaScited by 4
  2. 2025ALGEN: Few-shot Inversion Attacks on Textual Embeddings using Alignment and Generationcited by 4
  3. 2025RegionMarker: A Region-Triggered Semantic Watermarking Framework for Embedding-as-a-Service Copyright Protectioncited by 4
  4. 2024Protecting Copyright of Medical Pre-trained Language Models: Training-Free Backdoor Model Watermarkingcited by 0
  5. 2024Your Semantic-Independent Watermark is Fragile: A Semantic Perturbation Attack against EaaS Watermarkcited by 0