WaterMax: breaking the LLM watermark detectability-robustness-quality trade-off

arXiv:2403.04808v1 Announce Type: cross
Abstract: Watermarking is a technical means to dissuade malfeasant usage of Large Language Models. This paper proposes a novel watermarking scheme, so-called WaterMax, that enjoys high detectability while sustaining the quality of the generated text of the original LLM. Its new design leaves the LLM untouched (no modification of the weights, logits, temperature, or sampling technique). WaterMax balances robustness and complexity contrary to the watermarking techniques of the literature inherently provoking a trade-off between quality and …

abstract arxiv breaking cs.cl cs.cr cs.lg design generated language language models large language large language models llm novel paper quality robustness technical text trade trade-off type usage watermark watermarking