为扩散语言模型设计高效水印技术，实现低开销可靠检测。
LR-DWM: Efficient Watermarking for Diffusion Language Models
- 利用左右邻居信息动态调整生成令牌，适配非序列化扩散机制。
- 在标准评测下检测率超90%，运行时与内存开销接近无水印基线。
- 适合需要内容溯源的生成式AI应用，如论文、报告等场景。
水印是检测和归属人工智能生成内容的关键机制。当前大语言模型（LLM）的水印方法主要针对自回归（AR）模型：依赖文本逐个生成，并基于已采样文本嵌入稳定信号。扩散语言模型（DLM）通过非序列化的迭代去噪生成文本，难以直接应用传统水印方法。近期工作通过逆过程实现水印，但带来显著计算或内存开销。我们提出左-右扩散水印（LR-DWM），在可用时根据左右邻居偏差生成令牌。该方法运行时与内存开销极小，性能接近无水印基线，同时在标准评估设置下实现可靠统计检测。结果表明，DLM可被高效水印，达到高可检测性且开销可忽略。
原文摘要 · Abstract (English)
Watermarking (WM) is a critical mechanism for detecting and attributing AI-generated content. Current WM methods for Large Language Models (LLMs) are predominantly tailored for autoregressive (AR) models: They rely on tokens being generated sequentially, and embed stable signals within the generated sequence based on the previously sampled text. Diffusion Language Models (DLMs) generate text via non-sequential iterative denoising, which requires significant modification to use WM methods designed for AR models. Recent work proposed to watermark DLMs by inverting the process when needed, but suffers significant computational or memory overhead. We introduce Left-Right Diffusion Watermarking (LR-DWM), a scheme that biases the generated token based on both left and right neighbors, when they are available. LR-DWM incurs minimal runtime and memory overhead, remaining close to the non-watermarked baseline DLM while enabling reliable statistical detection under standard evaluation settings. Our results demonstrate that DLMs can be watermarked efficiently, achieving high detectability with negligible computational and memory overhead.
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。