WaterJudge: Quality-Detection Trade-off when Watermarking Large Language Models

arXiv:2403.19548v1 Announce Type: new
Abstract: Watermarking generative-AI systems, such as LLMs, has gained considerable interest, driven by their enhanced capabilities across a wide range of tasks. Although current approaches have demonstrated that small, context-dependent shifts in the word distributions can be used to apply and detect watermarks, there has been little work in analyzing the impact that these perturbations have on the quality of generated texts. Balancing high detectability with minimal performance degradation is crucial in terms of selecting the …

abstract ai systems apply arxiv capabilities context cs.cl current detection generative generative-ai language language models large language large language models llms quality small systems tasks trade trade-off type watermarking watermarks word work