一作ACL 2026Poster· Natural Language Processing
Q uantile M ark: A Message-Symmetric Multi-bit Watermark for LLM s
Junlin Zhu · Baizhou Huang · Xiaojun Wan
摘要
As large language models become standard backends for content generation, practical provenance increasingly requires multi-bit watermarking. In provider-internal deployments, a key requirement is message symmetry: the message itself should not systematically affect either text quality or verifica...