A Watermark for Low-entropy and Unbiased Generation in Large Language Models

• URL: http://arxiv.org/abs/2405.14604v2
• Date: Tue, 15 Oct 2024 20:14:50 GMT
• Title: A Watermark for Low-entropy and Unbiased Generation in Large Language Models
• Authors: Minjia Mao, Dongjun Wei, Zeyu Chen, Xiao Fang, Michael Chau,
• Abstract summary: This study proposes the Sampling One Then Accepting (STA-1) method, a watermark that can address all of these issues. We show that in low-entropy scenarios, unbiased watermarks face a tradeoff between watermark strength and the risk of unsatisfactory outputs. Experimental results on both low-entropy and high-entropy datasets demonstrate that STA-1 achieves text quality and watermark strength comparable to existing unbiased watermarks.
• Score: 6.505831742654826
• License:
• Abstract: Recent advancements in large language models (LLMs) have highlighted the risk of misusing them, raising the need for accurate detection of LLM-generated content. In response, a viable solution is to inject imperceptible identifiers into LLMs, known as watermarks. Previous work demonstrates that unbiased watermarks ensure unforgeability and preserve text quality by maintaining the expectation of the LLM output probability distribution. However, previous unbiased watermarking methods suffer from one or more of the following issues: (1) requiring access to white-box LLMs during detection, (2) incurring long detection time, (3) being not robust against simple watermarking attacks, (4) failing to provide statistical guarantees for the type II error of watermark detection, and (5) being not statistically unbiased for low-entropy scenarios, which hinder their deployment in practice. This study proposes the Sampling One Then Accepting (STA-1) method, a watermark that can address all of these issues. Moreover, we discuss the tradeoff between watermark strength and text quality for unbiased watermarks. We show that in low-entropy scenarios, unbiased watermarks face a tradeoff between watermark strength and the risk of unsatisfactory outputs. Experimental results on both low-entropy and high-entropy datasets demonstrate that STA-1 achieves text quality and watermark strength comparable to existing unbiased watermarks, with a low risk of unsatisfactory outputs. Implementation codes for this study are available online.

Related papers

• Can Watermarked LLMs be Identified by Users via Crafted Prompts? [55.460327393792156]
  This work is the first to investigate the imperceptibility of watermarked Large Language Models (LLMs) We design an identification algorithm called Water-Probe that detects watermarks through well-designed prompts. Experiments show that almost all mainstream watermarking algorithms are easily identified with our well-designed prompts.
  arXiv  Detail & Related papers  (2024-10-04T06:01:27Z)
• Distortion-free Watermarks are not Truly Distortion-free under Watermark Key Collisions [58.777395817878514]
  Language model (LM) watermarking techniques inject a statistical signal into LM-generated content. We introduce a new family of distortion-free watermarks--beta-watermark. Experimental results support that the beta-watermark can effectively reduce the distribution bias under key collisions.
  arXiv  Detail & Related papers  (2024-06-02T04:07:32Z)
• Black-Box Detection of Language Model Watermarks [1.9374282535132377]
  We develop rigorous statistical tests to detect the presence of all three most popular watermarking scheme families using only a limited number of black-box queries. Our findings indicate that current watermarking schemes are more detectable than previously believed, and that obscuring the fact that a watermark was deployed may not be a viable way for providers to protect against adversaries.
  arXiv  Detail & Related papers  (2024-05-28T08:41:30Z)
• Latent Watermark: Inject and Detect Watermarks in Latent Diffusion Space [7.082806239644562]
  Existing methods face the dilemma of image quality and watermark robustness. Watermarks with superior image quality usually have inferior robustness against attacks such as blurring and JPEG compression. We propose Latent Watermark, which injects and detects watermarks in the latent diffusion space.
  arXiv  Detail & Related papers  (2024-03-30T03:19:50Z)
• Duwak: Dual Watermarks in Large Language Models [49.00264962860555]
  We propose, Duwak, to enhance the efficiency and quality of watermarking by embedding dual secret patterns in both token probability distribution and sampling schemes. We evaluate Duwak extensively on Llama2, against four state-of-the-art watermarking techniques and combinations of them.
  arXiv  Detail & Related papers  (2024-03-12T16:25:38Z)
• WaterBench: Towards Holistic Evaluation of Watermarks for Large Language Models [48.19623266082828]
  WaterBench is the first comprehensive benchmark for watermarks in large language models (LLMs) We introduce WaterBench, the first comprehensive benchmark for LLM watermarks, in which we design three crucial factors. We evaluate $4$ open-source watermarks on $2$ LLMs under $2$ watermarking strengths and observe the common struggles for current methods on maintaining the generation quality.
  arXiv  Detail & Related papers  (2023-11-13T08:09:01Z)
• Unbiased Watermark for Large Language Models [67.43415395591221]
  This study examines how significantly watermarks impact the quality of model-generated outputs. It is possible to integrate watermarks without affecting the output probability distribution. The presence of watermarks does not compromise the performance of the model in downstream tasks.
  arXiv  Detail & Related papers  (2023-09-22T12:46:38Z)
• Provable Robust Watermarking for AI-Generated Text [41.5510809722375]
  We propose a robust and high-quality watermark method, Unigram-Watermark. We prove that our watermark method enjoys guaranteed generation quality, correctness in watermark detection, and is robust against text editing and paraphrasing.
  arXiv  Detail & Related papers  (2023-06-30T07:24:32Z)
• On the Reliability of Watermarks for Large Language Models [95.87476978352659]
  We study the robustness of watermarked text after it is re-written by humans, paraphrased by a non-watermarked LLM, or mixed into a longer hand-written document. We find that watermarks remain detectable even after human and machine paraphrasing. We also consider a range of new detection schemes that are sensitive to short spans of watermarked text embedded inside a large document.
  arXiv  Detail & Related papers  (2023-06-07T17:58:48Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.