Unbiased Watermark for Large Language Models. (arXiv:2310.10669v2 [cs.CR] UPDATED)

The recent advancements in large language models (LLMs) have sparked a
growing apprehension regarding the potential misuse. One approach to mitigating
this risk is to incorporate watermarking techniques into LLMs, allowing for the
tracking and attribution of model outputs. This study examines a crucial aspect
of watermarking: how significantly watermarks impact the quality of
model-generated outputs. Previous studies have suggested a trade-off between
watermark strength and output quality. However, our research demonstrates that
it is possible to integrate watermarks without …

aspect attribution impact language language models large llms risk study techniques tracking watermarking watermarks