# PostMark: A Robust Blackbox Watermark for Large Language Models Source: [PostMark: A Robust Blackbox Watermark for Large Language Models]() Collection index: [llms.txt]() Data updated: 2025-10-26T03:28:08.430Z Authors: Yapei Chang, Kalpesh Krishna, Amir Houmansadr, John Wieting, Mohit Iyyer Published: 2024-06-20T17:27:14.000Z ## Abstract The most effective techniques to detect LLM-generated text rely on inserting a detectable signature -- or watermark -- during the model's decoding process. Most existing watermarking methods require access to the underlying LLM's logits, which LLM API providers are loath to share due to fears of model distillation. As such, these watermarks must be implemented independently by each LLM provider. In this paper, we develop PostMark, a modular post-hoc watermarking procedure in which an input-dependent set of words (determined via a semantic embedding) is inserted into the text after the decoding process has completed. Critically, PostMark does not require logit access, which means it can be implemented by a third party. We also show that PostMark is more robust to paraphrasing attacks than existing watermarking methods: our experiments cover eight baseline algorithms, five base LLMs, and three datasets. Finally, we evaluate the impact of PostMark on text quality using both automated and human assessments, highlighting the trade-off between quality and robustness to paraphrasing. We release our code, outputs, and annotations at https://github.com/lilakk/PostMark. ## Recorded benchmark results A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. - [Watermarking Robustness on DTS Deletion - 30%](): Watermarking Robustness; DTS Deletion - 30%; ROC-AUC: 0.908. - [Watermarking Robustness on DTS Replacement 30%](): Watermarking Robustness; DTS Replacement 30%; ROC-AUC: 95.6. - [Watermarking Robustness on DTS Paraphrase - ChatGPT](): Watermarking Robustness; DTS Paraphrase - ChatGPT; ROC-AUC: 0.841. - [Watermark Detection on ELI5 and ICLR papers DTS setting](): Watermark Detection; ELI5 and ICLR papers DTS setting; ROC-AUC: 97.7. ## Original sources and implementation [Original paper]() [PDF]()