为离散扩散语言模型设计了无损水印技术，可高效追踪生成内容。
Watermarking Discrete Diffusion Language Models
- 在每步扩散中使用保持分布的Gumbel-max采样，结合位置种子控制随机性。
- 在LLaDA模型上实现可靠检测，错误检测率随序列长度指数下降。
- 无需调参即可部署，适合大规模模型和跨任务应用。
水印技术已成为追踪AI生成内容、区分其与人工创作的重要手段。尽管已有大量研究针对自回归大语言模型（LLMs）和图像扩散模型的水印，但对因推理吞吐高而日益流行的离散扩散语言模型（DDLMs）的水印研究仍相对不足。本文提出首个针对DDLMs的水印方法：在每一步扩散过程中采用保持分布的Gumbel-max采样，并以序列位置作为随机性种子，从而实现可靠检测。我们在当前最先进的DDLM——LLaDA上实证验证了该方法的有效性。理论分析证明，该水印为无失真设计，且错误检测概率随序列长度呈指数级衰减。一个关键实际优势是，该方法无需昂贵的超参数调优，可直接部署并扩展至多种模型与基准测试。
原文摘要 · Abstract (English)
Watermarking has emerged as a promising technique to track AI-generated content and differentiate it from authentic human creations. While prior work extensively studies watermarking for autoregressive large language models (LLMs) and image diffusion models, it remains comparatively underexplored for discrete diffusion language models (DDLMs), which are becoming popular due to their high inference throughput. In this paper, we introduce one of the first watermarking methods for DDLMs. Our approach applies a distribution-preserving Gumbel-max sampling trick at every diffusion step and seeds the randomness by sequence position to enable reliable detection. We empirically demonstrate reliable detectability on LLaDA, a state-of-the-art DDLM. We also analytically prove that the watermark is distortion-free, with a false detection probability that decays exponentially in the sequence length. A key practical advantage is that our method realizes desired watermarking properties with no expensive hyperparameter tuning, making it straightforward to deploy and scale across models and benchmarks.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。