Token-Specific Watermarking with Enhanced Detectability and Semantic Coherence for Large Language Models

arXiv:2402.18059v1 Announce Type: cross
Abstract: Large language models generate high-quality responses with potential misinformation, underscoring the need for regulation by distinguishing AI-generated and human-written texts. Watermarking is pivotal in this context, which involves embedding hidden markers in texts during the LLM inference phase, which is imperceptible to humans. Current watermarking algorithms, however, face the challenge of achieving both the detectability of inserted watermarks and the semantic integrity of generated texts, where enhancing one aspect often undermines the other. To overcome …

arxiv cs.cl cs.cr cs.lg language language models large semantic token watermarking