Towards Better Statistical Understanding of Watermarking LLMs
In this paper, we study the problem of watermarking large language models (LLMs). We consider the trade-off between model distortion and detection ability and formulate it as a constrained optimization problem based on the green-red algorithm of Kirchenbauer et al. (2023a). We show that the optimal solution to the optimization problem enjoys a nice analytical property which provides a better understanding and inspires the algorithm design for the watermarking process. We develop an online dual gradient ascent watermarking algorithm in light of this optimization formulation and prove its asymptotic Pareto optimality between model distortion and detection ability. Such a result guarantees an averaged increased green list probability and henceforth detection ability explicitly (in contrast to previous results). Moreover, we provide a systematic discussion on the choice of the model distortion metrics for the watermarking problem. We justify our choice of KL divergence and present issues with the existing criteria of ``distortion-free'' and perplexity. Finally, we empirically evaluate our algorithms on extensive datasets against benchmark algorithms.
Code (1)
Similar Papers 제목 키워드 기반
SimMark: A Robust Sentence-Level Similarity-Based Watermarking Algorithm for Large Language Models
The rapid proliferation of large language models (LLMs) has created an urgent need for reliable methods to detect whether a text is generated by such models. In this paper, we propose SimMark, a posthoc watermarking algo…
SentenceSentence EmbeddingsA Survey of Text Watermarking in the Era of Large Language Models
Text watermarking algorithms are crucial for protecting the copyright of textual content. Historically, their capabilities and application scenarios were limited. However, recent advancements in large language models (LL…
Dialogue GenerationSurveyText GenerationBlind PRNG Hijacking: An Undetectable Integrity-Preserving Attack Against LLM Watermarking
Cryptographic watermarking is a leading defense for attributing text generated by large language models (LLMs). Existing schemes, including KGW, Unigram, and DipMark, derive their security guarantees from the assumption …
Waterfall: Framework for Robust and Scalable Text Watermarking and Provenance for LLMs
Protecting intellectual property (IP) of text such as articles and code is increasingly important, especially as sophisticated attacks become possible, such as paraphrasing by large language models (LLMs) or even unautho…
ArticlesComputational EfficiencyLLM Output Detectability and Task Performance Can be Jointly Optimized
Detecting machine-generated text is essential for transparency and accountability when deploying LLMs. Watermarking enables statistically reliable detection by biasing token distributions to embed detectable signals into…
Reinforcement Learning