MC$^2$Mark: Distortion-Free Multi-Bit Watermarking for Long Messages
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2026
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866915799533355008 |
|---|---|
| author | Cui, Xuehao Chen, Ruibo Wu, Yihan Huang, Heng |
| author_facet | Cui, Xuehao Chen, Ruibo Wu, Yihan Huang, Heng |
| contents | Large language models now produce text indistinguishable from human writing, which increases the need for reliable provenance tracing. Multi-bit watermarking can embed identifiers into generated text, but existing methods struggle to keep both text quality and watermark strength while carrying long messages. We propose MC$^2$Mark, a distortion-free multi-bit watermarking framework designed for reliable embedding and decoding of long messages. Our key technical idea is Multi-Channel Colored Reweighting, which encodes bits through structured token reweighting while keeping the token distribution unbiased, together with Multi-Layer Sequential Reweighting to strengthen the watermark signal and an evidence-accumulation detector for message recovery. Experiments show that MC$^2$Mark improves detectability and robustness over prior multi-bit watermarking methods while preserving generation quality, achieving near-perfect accuracy for short messages and exceeding the second-best method by nearly 30% for long messages. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_14030 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | MC$^2$Mark: Distortion-Free Multi-Bit Watermarking for Long Messages Cui, Xuehao Chen, Ruibo Wu, Yihan Huang, Heng Cryptography and Security Machine Learning Large language models now produce text indistinguishable from human writing, which increases the need for reliable provenance tracing. Multi-bit watermarking can embed identifiers into generated text, but existing methods struggle to keep both text quality and watermark strength while carrying long messages. We propose MC$^2$Mark, a distortion-free multi-bit watermarking framework designed for reliable embedding and decoding of long messages. Our key technical idea is Multi-Channel Colored Reweighting, which encodes bits through structured token reweighting while keeping the token distribution unbiased, together with Multi-Layer Sequential Reweighting to strengthen the watermark signal and an evidence-accumulation detector for message recovery. Experiments show that MC$^2$Mark improves detectability and robustness over prior multi-bit watermarking methods while preserving generation quality, achieving near-perfect accuracy for short messages and exceeding the second-best method by nearly 30% for long messages. |
| title | MC$^2$Mark: Distortion-Free Multi-Bit Watermarking for Long Messages |
| topic | Cryptography and Security Machine Learning |
| url | https://arxiv.org/abs/2602.14030 |